nips nips2012 nips2012-9 knowledge-graph by maker-knowledge-mining

9 nips-2012-A Geometric take on Metric Learning

Source: pdf

Author: Søren Hauberg, Oren Freifeld, Michael J. Black

Abstract: Multi-metric learning techniques learn local metric tensors in different parts of a feature space. With such an approach, even simple classiﬁers can be competitive with the state-of-the-art because the distance measure locally adapts to the structure of the data. The learned distance measure is, however, non-metric, which has prevented multi-metric learning from generalizing to tasks such as dimensionality reduction and regression in a principled way. We prove that, with appropriate changes, multi-metric learning corresponds to learning the structure of a Riemannian manifold. We then show that this structure gives us a principled way to perform dimensionality reduction and regression according to the learned metrics. Algorithmically, we provide the ﬁrst practical algorithm for computing geodesics according to the learned metrics, as well as algorithms for computing exponential and logarithmic maps on the Riemannian manifold. Together, these tools let many Euclidean algorithms take advantage of multi-metric learning. We illustrate the approach on regression and dimensionality reduction tasks that involve predicting measurements of the human body from shape data. 1 Learning and Computing Distances Statistics relies on measuring distances. When the Euclidean metric is insufﬁcient, as is the case in many real problems, standard methods break down. This is a key motivation behind metric learning, which strives to learn good distance measures from data. In the most simple scenarios a single metric tensor is learned, but in recent years, several methods have proposed learning multiple metric tensors, such that different distance measures are applied in different parts of the feature space. This has proven to be a very powerful approach for classiﬁcation tasks [1, 2], but the approach has not generalized to other tasks. Here we consider the generalization of Principal Component Analysis (PCA) and linear regression; see Fig. 1 for an illustration of our approach. The main problem with generalizing multi-metric learning is that it is based on assumptions that make the feature space both non-smooth and non-metric. Speciﬁcally, it is often assumed that straight lines form geodesic curves and that the metric tensor stays constant along these lines. These assumptions are made because it is believed that computing the actual geodesics is intractable, requiring a discretization of the entire feature space [3]. We solve these problems by smoothing the transitions between different metric tensors, which ensures a metric space where geodesics can be computed. In this paper, we consider the scenario where the metric tensor at a given point in feature space is deﬁned as the weighted average of a set of learned metric tensors. In this model, we prove that the feature space becomes a chart for a Riemannian manifold. This ensures a metric feature space, i.e. dist(x, y) = 0 ⇔ x = y , dist(x, y) = dist(y, x) (symmetry), (1) dist(x, z) ≤ dist(x, y) + dist(y, z) (triangle inequality). To compute statistics according to the learned metric, we need to be able to compute distances, which implies that we need to compute geodesics. Based on the observation that geodesics are 1 (a) Local Metrics & Geodesics (b) Tangent Space Representation (c) First Principal Geodesic Figure 1: Illustration of Principal Geodesic Analysis. (a) Geodesics are computed between the mean and each data point. (b) Data is mapped to the Euclidean tangent space and the ﬁrst principal component is computed. (c) The principal component is mapped back to the feature space. smooth curves in Riemannian spaces, we derive an algorithm for computing geodesics that only requires a discretization of the geodesic rather than the entire feature space. Furthermore, we show how to compute the exponential and logarithmic maps of the manifold. With this we can map any point back and forth between a Euclidean tangent space and the manifold. This gives us a general strategy for incorporating the learned metric tensors in many Euclidean algorithms: map the data to the tangent of the manifold, perform the Euclidean analysis and map the results back to the manifold. Before deriving the algorithms (Sec. 3) we set the scene by an analysis of the shortcomings of current state-of-the-art methods (Sec. 2), which motivate our ﬁnal model. The model is general and can be used for many problems. Here we illustrate it with several challenging problems in 3D body shape modeling and analysis (Sec. 4). All proofs can be found in the supplementary material along with algorithmic details and further experimental results. 2 Background and Related Work Single-metric learning learns a metric tensor, M, such that distances are measured as dist2 (xi , xj ) = xi − xj 2 M ≡ (xi − xj )T M(xi − xj ) , (2) where M is a symmetric and positive deﬁnite D × D matrix. Classic approaches for ﬁnding such a metric tensor include PCA, where the metric is given by the inverse covariance matrix of the training data; and linear discriminant analysis (LDA), where the metric tensor is M = S−1 SB S−1 , with Sw W W and SB being the within class scatter and the between class scatter respectively [9]. A more recent approach tries to learn a metric tensor from triplets of data points (xi , xj , xk ), where the metric should obey the constraint that dist(xi , xj ) < dist(xi , xk ). Here the constraints are often chosen such that xi and xj belong to the same class, while xi and xk do not. Various relaxed versions of this idea have been suggested such that the metric can be learned by solving a semi-deﬁnite or a quadratic program [1, 2, 4–8]. Among the most popular approaches is the Large Margin Nearest Neighbor (LMNN) classiﬁer [5], which ﬁnds a linear transformation that satisﬁes local distance constraints, making the approach suitable for multi-modal classes. For many problems, a single global metric tensor is not enough, which motivates learning several local metric tensors. The classic work by Hastie and Tibshirani [9] advocates locally learning metric tensors according to LDA and using these as part of a kNN classiﬁer. In a somewhat similar fashion, Weinberger and Saul [5] cluster the training data and learn a separate metric tensor for each cluster using LMNN. A more extreme point of view was taken by Frome et al. [1, 2], who learn a diagonal metric tensor for every point in the training set, such that distance rankings are preserved. Similarly, Malisiewicz and Efros [6] ﬁnd a diagonal metric tensor for each training point such that the distance to a subset of the training data from the same class is kept small. Once a set of metric tensors {M1 , . . . , MR } has been learned, the distance dist(a, b) is measured according to (2) where “the nearest” metric tensor is used, i.e. R M(x) = r=1 wr (x) ˜ Mr , where wr (x) = ˜ ˜ j wj (x) 1 0 x − xr 2 r ≤ x − xj M otherwise 2 Mj , ∀j , (3) where x is either a or b depending on the algorithm. Note that this gives a non-metric distance function as it is not symmetric. To derive this equation, it is necessary to assume that 1) geodesics 2 −8 −8 Assumed Geodesics Location of Metric Tensors Test Points −6 −8 Actual Geodesics Location of Metric Tensors Test Points −6 Riemannian Geodesics Location of Metric Tensors Test Points −6 −4 −4 −4 −2 −2 −2 0 0 0 2 2 2 4 4 4 6 −8 6 −8 −6 −4 −2 0 (a) 2 4 6 −6 −4 −2 0 2 4 6 6 −8 −6 (b) −4 −2 (c) 0 2 4 6 (d) Figure 2: (a)–(b) An illustrative example where straight lines do not form geodesics and where the metric tensor does not stay constant along lines; see text for details. The background color is proportional to the trace of the metric tensor, such that light grey corresponds to regions where paths are short (M1 ), and dark grey corresponds to regions they are long (M2 ). (c) The suggested geometric model along with the geodesics. Again, background colour is proportional to the trace of the metric tensor; the colour scale is the same is used in (a) and (b). (d) An illustration of the exponential and logarithmic maps. form straight lines, and 2) the metric tensor stays constant along these lines [3]. Both assumptions are problematic, which we illustrate with a simple example in Fig. 2a–c. Assume we are given two metric tensors M1 = 2I and M2 = I positioned at x1 = (2, 2)T and x2 = (4, 4)T respectively. This gives rise to two regions in feature space in which x1 is nearest in the ﬁrst and x2 is nearest in the second, according to (3). This is illustrated in Fig. 2a. In the same ﬁgure, we also show the assumed straight-line geodesics between selected points in space. As can be seen, two of the lines goes through both regions, such that the assumption of constant metric tensors along the line is violated. Hence, it would seem natural to measure the length of the line, by adding the length of the line segments which pass through the different regions of feature space. This was suggested by Ramanan and Baker [3] who also proposed a polynomial time algorithm for measuring these line lengths. This gives a symmetric distance function. Properly computing line lengths according to the local metrics is, however, not enough to ensure that the distance function is metric. As can be seen in Fig. 2a the straight line does not form a geodesic as a shorter path can be found by circumventing the region with the “expensive” metric tensor M1 as illustrated in Fig. 2b. This issue makes it trivial to construct cases where the triangle inequality is violated, which again makes the line length measure non-metric. In summary, if we want a metric feature space, we can neither assume that geodesics are straight lines nor that the metric tensor stays constant along such lines. In practice, good results have been reported using (3) [1,3,5], so it seems obvious to ask: is metricity required? For kNN classiﬁers this does not appear to be the case, with many successes based on dissimilarities rather than distances [10]. We, however, want to generalize PCA and linear regression, which both seek to minimize the reconstruction error of points projected onto a subspace. As the notion of projection is hard to deﬁne sensibly in non-metric spaces, we consider metricity essential. In order to build a model with a metric feature space, we change the weights in (3) to be smooth functions. This impose a well-behaved geometric structure on the feature space, which we take advantage of in order to perform statistical analysis according to the learned metrics. However, ﬁrst we review the basics of Riemannian geometry as this provides the theoretical foundation of our work. 2.1 Geodesics and Riemannian Geometry We start by deﬁning Riemannian manifolds, which intuitively are smoothly curved spaces equipped with an inner product. Formally, they are smooth manifolds endowed with a Riemannian metric [11]: Deﬁnition A Riemannian metric M on a manifold M is a smoothly varying inner product < a, b >x = aT M(x)b in the tangent space Tx M of each point x ∈ M . 3 Often Riemannian manifolds are represented by a chart; i.e. a parameter space for the curved surface. An example chart is the spherical coordinate system often used to represent spheres. While such charts are often ﬂat spaces, the curvature of the manifold arises from the smooth changes in the metric. On a Riemannian manifold M, the length of a smooth curve c : [0, 1] → M is deﬁned as the integral of the norm of the tangent vector (interpreted as speed) along the curve: 1 Length(c) = 1 c (λ) M(c(λ)) dλ c (λ)T M(c(λ))c (λ)dλ , = (4) 0 0 where c denotes the derivative of c and M(c(λ)) is the metric tensor at c(λ). A geodesic curve is then a length-minimizing curve connecting two given points x and y, i.e. (5) cgeo = arg min Length(c) with c(0) = x and c(1) = y . c The distance between x and y is deﬁned as the length of the geodesic. Given a tangent vector v ∈ Tx M, there exists a unique geodesic cv (t) with initial velocity v at x. The Riemannian exponential map, Expx , maps v to a point on the manifold along the geodesic cv at t = 1. This mapping preserves distances such that dist(cv (0), cv (1)) = v . The inverse of the exponential map is the Riemannian logarithmic map denoted Logx . Informally, the exponential and logarithmic maps move points back and forth between the manifold and the tangent space while preserving distances (see Fig. 2d for an illustration). This provides a general strategy for generalizing many Euclidean techniques to Riemannian domains: data points are mapped to the tangent space, where ordinary Euclidean techniques are applied and the results are mapped back to the manifold. 3 A Metric Feature Space With the preliminaries settled we deﬁne the new model. Let C = RD denote the feature space. We endow C with a metric tensor in every point x, which we deﬁne akin to (3), R M(x) = wr (x)Mr , where wr (x) = r=1 wr (x) ˜ R ˜ j=1 wj (x) , (6) with wr > 0. The only difference from (3) is that we shall not restrict ourselves to binary weight ˜ functions wr . We assume the metric tensors Mr have already been learned; Sec. 4 contain examples ˜ where they have been learned using LMNN [5] and LDA [9]. From the deﬁnition of a Riemannian metric, we trivially have the following result: Lemma 1 The space C = RD endowed with the metric tensor from (6) is a chart of a Riemannian manifold, iff the weights wr (x) change smoothly with x. Hence, by only considering smooth weight functions wr we get a well-studied geometric structure ˜ on the feature space, which ensures us that it is metric. To illustrate the implications we return to the example in Fig. 2. We change the weight functions from binary to squared exponentials, which gives the feature space shown in Fig. 2c. As can be seen, the metric tensor now changes smoothly, which also makes the geodesics smooth curves (a property we will use when computing the geodesics). It is worth noting that Ramanan and Baker [3] also consider the idea of smoothly averaging the metric tensor. They, however, only evaluate the metric tensor at the test point of their classiﬁer and then assume straight line geodesics with a constant metric tensor. Such assumptions violate the premise of a smoothly changing metric tensor and, again, the distance measure becomes non-metric. Lemma 1 shows that metric learning can be viewed as manifold learning. The main difference between our approach and techniques such as Isomap [12] is that, while Isomap learns an embedding of the data points, we learn the actual manifold structure. This gives us the beneﬁt that we can compute geodesics as well as the exponential and logarithmic maps. These provide us with mappings back and forth between the manifold and Euclidean representation of the data, which preserve distances as well as possible. The availability of such mappings is in stark contrast to e.g. Isomap. In the next section we will derive a system of ordinary differential equations (ODE’s) that geodesics in C have to satisfy, which provides us with algorithms for computing geodesics as well as exponential and logarithmic maps. With these we can generalize many Euclidean techniques. 4 3.1 Computing Geodesics, Maps and Statistics At minima of (4) we know that the Euler-Lagrange equation must hold [11], i.e. ∂L d ∂L , where L(λ, c, c ) = c (λ)T M(c(λ))c (λ) . = ∂c dλ ∂c As we have an explicit expression for the metric tensor we can compute (7) in closed form: (7) Theorem 2 Geodesic curves in C satisfy the following system of 2nd order ODE’s M(c(λ))c (λ) = − 1 ∂vec [M(c(λ))] 2 ∂c(λ) T (c (λ) ⊗ c (λ)) , (8) where ⊗ denotes the Kronecker product and vec [·] stacks the columns of a matrix into a vector [13]. Proof See supplementary material. This result holds for any smooth weight functions wr . We, however, still need to compute ∂vec[M] , ˜ ∂c which depends on the speciﬁc choice of wr . Any smooth weighting scheme is applicable, but we ˜ restrict ourselves to the obvious smooth generalization of (3) and use squared exponentials. From this assumption, we get the following result Theorem 3 For wr (x) = exp − ρ x − xr ˜ 2 ∂vec [M(c)] = ∂c the derivative of the metric tensor from (6) is R ρ R j=1 2 Mr R 2 wj ˜ T r=1 T wj (c − xj ) Mj − (c − xr ) Mr ˜ wr vec [Mr ] ˜ . (9) j=1 Proof See supplementary material. Computing Geodesics. Any geodesic curve must be a solution to (8). Hence, to compute a geodesic between x and y, we can solve (8) subject to the constraints c(0) = x and c(1) = y . (10) This is a boundary value problem, which has a smooth solution. This allows us to solve the problem numerically using a standard three-stage Lobatto IIIa formula, which provides a fourth-order accurate C 1 –continuous solution [14]. Ramanan and Baker [3] discuss the possibility of computing geodesics, but arrive at the conclusion that this is intractable based on the assumption that it requires discretizing the entire feature space. Our solution avoids discretizing the feature space by discretizing the geodesic curve instead. As this is always one-dimensional the approach remains tractable in high-dimensional feature spaces. Computing Logarithmic Maps. Once a geodesic c is found, it follows from the deﬁnition of the logarithmic map, Logx (y), that it can be computed as v = Logx (y) = c (0) Length(c) . c (0) (11) In practice, we solve (8) by rewriting it as a system of ﬁrst order ODE’s, such that we compute both c and c simultaneously (see supplementary material for details). Computing Exponential Maps. Given a starting point x on the manifold and a vector v in the tangent space, the exponential map, Expx (v), ﬁnds the unique geodesic starting at x with initial velocity v. As the geodesic must fulﬁll (8), we can compute the exponential map by solving this system of ODE’s with the initial conditions c(0) = x and c (0) = v . (12) This initial value problem has a unique solution, which we ﬁnd numerically using a standard RungeKutta scheme [15]. 5 3.1.1 Generalizing PCA and Regression At this stage, we know that the feature space is Riemannian and we know how to compute geodesics and exponential and logarithmic maps. We now seek to generalize PCA and linear regression, which becomes straightforward since solutions are available in Riemannian spaces [16, 17]. These generalizations can be summarized as mapping the data to the tangent space at the mean, performing standard Euclidean analysis in the tangent and mapping the results back. The ﬁrst step is to compute the mean value on the manifold, which is deﬁned as the point that minimizes the sum-of-squares distances to the data points. Pennec [18] provides an efﬁcient gradient descent approach for computing this point, which we also summarize in the supplementary material. The empirical covariance of a set of points is deﬁned as the ordinary Euclidean covariance in the tangent space at the mean value [18]. With this in mind, it is not surprising that the principal components of a dataset have been generalized as the geodesics starting at the mean with initial velocity corresponding to the eigenvectors of the covariance [16], γvd (t) = Expµ (tvd ) , (13) th where vd denotes the d eigenvector of the covariance. This approach is called Principal Geodesic Analysis (PGA), and the geodesic curve γvd is called the principal geodesic. An illustration of the approach can be seen in Fig. 1 and more algorithmic details are in the supplementary material. Linear regression has been generalized in a similar way [17] by performing regression in the tangent of the mean and mapping the resulting line back to the manifold using the exponential map. The idea of working in the tangent space is both efﬁcient and convenient, but comes with an element of approximation as the logarithmic map is only guarantied to preserve distances to the origin of the tangent and not between all pairs of data points. Practical experience, however, indicates that this is a good tradeoff; see [19] for a more in-depth discussion of when the approximation is suitable. 4 Experiments To illustrate the framework1 we consider an example in human body analysis, and then we analyze the scalability of the approach. But ﬁrst, to build intuition, Fig. 3a show synthetically generated data samples from two classes. We sample random points xr and learn a local LDA metric [9] by considering all data points within a radius; this locally pushes the two classes apart. We combine the local metrics using (6) and Fig. 3b show the data in the tangent space of the resulting manifold. As can be seen the two classes are now globally further apart, which shows the effect of local metrics. 4.1 Human Body Shape We consider a regression example concerning human body shape analysis. We study 986 female body laser scans from the CAESAR [20] data set; each shape is represented using the leading 35 principal components of the data learned using a SCAPE-like model [21, 22]. Each shape is associated with anthropometric measurements such as body height, shoe size, etc. We show results for shoulder to wrist distance and shoulder breadth, but results for more measurements are in the supplementary material. To predict the measurements from shape coefﬁcients, we learn local metrics and perform linear regression according to these. As a further experiment, we use PGA to reduce the dimensionality of the shape coefﬁcients according to the local metrics, and measure the quality of the reduction by performing linear regression to predict the measurements. As a baseline we use the corresponding Euclidean techniques. To learn the local metric we do the following. First we whiten the data such that the variance captured by PGA will only be due to the change of metric; this allows easy visualization of the impact of the learned metrics. We then cluster the body shapes into equal-sized clusters according to the measurement and learn a LMNN metric for each cluster [5], which we associate with the mean of each class. These push the clusters apart, which introduces variance along the directions where the measurement changes. From this we construct a Riemannian manifold according to (6), 1 Our software implementation for computing geodesics and performing manifold statistics is available at http://ps.is.tue.mpg.de/project/Smooth Metric Learning 6 30 Euclidean Model Riemannian Model 24 20 18 16 20 15 10 5 14 12 0 (a) 25 22 Running Time (sec.) Average Prediction Error 26 10 (b) 20 Dimensionality 0 0 30 50 (c) 100 Dimensionality 150 (d) 4 3 3 2 2 1 1 0 −1 −2 −3 −4 −4 −3 −2 −1 0 1 2 3 4 Shoulder breadth 20 −2 −3 Euclidean Model Riemannian Model 0 −1 25 Prediction Error 4 15 10 0 −4 −5 0 4 10 15 20 Dimensionality 16 25 30 35 17 3 3 5 5 Euclidean Model Riemannian Model 2 15 2 1 1 Prediction Error Shoulder to wrist distance Figure 3: Left panels: Synthetic data. (a) Samples from two classes along with illustratively sampled metric tensors from (6). (b) The data represented in the tangent of a manifold constructed from local LDA metrics learned at random positions. Right panels: Real data. (c) Average error of linearly predicted body measurements (mm). (d) Running time (sec) of the geodesic computation as a function of dimensionality. 0 0 −1 −2 −1 −3 14 13 12 11 −2 −4 −3 −4 −4 10 −5 −3 −2 −1 0 1 Euclidean PCA 2 3 −6 −4 9 0 −2 0 2 4 Tangent Space PCA (PGA) 6 5 10 15 20 Dimensionality 25 30 35 Regression Error Figure 4: Left: body shape data in the ﬁrst two principal components according to the Euclidean metric. Point color indicates cluster membership. Center: As on the left, but according to the Riemannian model. Right: regression error as a function of the dimensionality of the shape space; again the Euclidean metric and the Riemannian metric are compared. compute the mean value on the manifold, map the data to the tangent space at the mean and perform linear regression in the tangent space. As a ﬁrst visualization we plot the data expressed in the leading two dimensions of PGA in Fig. 4; as can be seen the learned metrics provide principal geodesics, which are more strongly related with the measurements than the Euclidean model. In order to predict the measurements from the body shape, we perform linear regression, both directly in the shape space according to the Euclidean metric and in the tangent space of the manifold corresponding to the learned metrics (using the logarithmic map from (11)). We measure the prediction error using leave-one-out cross-validation. To further illustrate the power of the PGA model, we repeat this experiment for different dimensionalities of the data. The results are plotted in Fig. 4, showing that regression according to the learned metrics outperforms the Euclidean model. To verify that the learned metrics improve accuracy, we average the prediction errors over all millimeter measurements. The result in Fig. 3c shows that much can be gained in lower dimensions by using the local metrics. To provide visual insights into the behavior of the learned metrics, we uniformly sample body shape along the ﬁrst principal geodesic (in the range ±7 times the standard deviation) according to the different metrics. The results are available as a movie in the supplementary material, but are also shown in Fig. 5. As can be seen, the learned metrics pick up intuitive relationships between body shape and the measurements, e.g. shoulder to wrist distance is related to overall body size, while shoulder breadth is related to body weight. 7 Shoulder to wrist distance Shoulder breadth Figure 5: Shapes corresponding to the mean (center) and ±7 times the standard deviations along the principal geodesics (left and right). Movies are available in the supplementary material. 4.2 Scalability The human body data set is small enough (986 samples in 35 dimensions) that computing a geodesic only takes a few seconds. To show that the current unoptimized Matlab implementation can handle somewhat larger datasets, we brieﬂy consider a dimensionality reduction task on the classic MNIST handwritten digit data set. We use the preprocessed data available with [3] where the original 28×28 gray scale images were deskewed and projected onto their leading 164 Euclidean principal components (which captures 95% of the variance in the original data). We learn one diagonal LMNN metric per class, which we associate with the mean of the class. From this we construct a Riemannian manifold from (6), compute the mean value on the manifold and compute geodesics between the mean and each data point; this is the computationally expensive part of performing PGA. Fig. 3d plots the average running time (sec) for the computation of geodesics as a function of the dimensionality of the training data. A geodesic can be computed in 100 dimensions in approximately 5 sec., whereas in 150 dimensions it takes about 30 sec. In this experiment, we train a PGA model on 60,000 data points, and test a nearest neighbor classiﬁer in the tangent space as we decrease the dimensionality of the model. Compared to a Euclidean model, this gives a modest improvement in classiﬁcation accuracy of 2.3 percent, when averaged across different dimensionalities. Plots of the results can be found in the supplementary material. 5 Discussion This work shows that multi-metric learning techniques are indeed applicable outside the realm of kNN classiﬁers. The idea of deﬁning the metric tensor at any given point as the weighted average of a ﬁnite set of learned metrics is quite natural from a modeling point of view, which is also validated by the Riemannian structure of the resulting space. This opens both a theoretical and a practical toolbox for analyzing and developing algorithms that use local metric tensors. Speciﬁcally, we show how to use local metric tensors for both regression and dimensionality reduction tasks. Others have attempted to solve non-classiﬁcation problems using local metrics, but we feel that our approach is the ﬁrst to have a solid theoretical backing. For example, Hastie and Tibshirani [9] use local LDA metrics for dimensionality reduction by averaging the local metrics and using the resulting metric as part of a Euclidean PCA, which essentially is a linear approach. Another approach was suggested by Hong et al. [23] who simply compute the principal components according to each metric separately, such that one low dimensional model is learned per metric. The suggested approach is, however, not difﬁculty-free in its current implementation. Currently, we are using off-the-shelf numerical solvers for computing geodesics, which can be computationally demanding. While we managed to analyze medium-sized datasets, we believe that the run-time can be drastically improved by developing specialized numerical solvers. In the experiments, we learned local metrics using techniques specialized for classiﬁcation tasks as this is all the current literature provides. We expect improvements by learning the metrics speciﬁcally for regression and dimensionality reduction, but doing so is currently an open problem. Acknowledgments: Søren Hauberg is supported in part by the Villum Foundation, and Oren Freifeld is supported in part by NIH-NINDS EUREKA (R01-NS066311). 8 References [1] Andrea Frome, Yoram Singer, and Jitendra Malik. Image retrieval and classiﬁcation using local distance functions. In B. Sch¨ lkopf, J. Platt, and T. Hoffman, editors, Advances in Neural Information Processing o Systems 19 (NIPS), pages 417–424, Cambridge, MA, 2007. MIT Press. [2] Andrea Frome, Fei Sha, Yoram Singer, and Jitendra Malik. Learning globally-consistent local distance functions for shape-based image retrieval and classiﬁcation. In International Conference on Computer Vision (ICCV), pages 1–8, 2007. [3] Deva Ramanan and Simon Baker. Local distance functions: A taxonomy, new algorithms, and an evaluation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 33(4):794–806, 2011. [4] Shai Shalev-Shwartz, Yoram Singer, and Andrew Y. Ng. Online and batch learning of pseudo-metrics. In Proceedings of the twenty-ﬁrst international conference on Machine learning, ICML ’04, pages 94–101. ACM, 2004. [5] Kilian Q. Weinberger and Lawrence K. Saul. Distance metric learning for large margin nearest neighbor classiﬁcation. The Journal of Machine Learning Research, 10:207–244, 2009. [6] Tomasz Malisiewicz and Alexei A. Efros. Recognition by association via learning per-exemplar distances. In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 1–8, 2008. [7] Yiming Ying and Peng Li. Distance metric learning with eigenvalue optimization. The Journal of Machine Learning Research, 13:1–26, 2012. [8] Matthew Schultz and Thorsten Joachims. Learning a distance metric from relative comparisons. In Advances in Neural Information Processing Systems 16 (NIPS), 2004. [9] Trevor Hastie and Robert Tibshirani. Discriminant adaptive nearest neighbor classiﬁcation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 18(6):607–616, June 1996. [10] Elzbieta Pekalska, Pavel Paclik, and Robert P. W. Duin. A generalized kernel approach to dissimilaritybased classiﬁcation. Journal of Machine Learning Research, 2:175–211, 2002. [11] Manfredo Perdigao do Carmo. Riemannian Geometry. Birkh¨ user Boston, January 1992. a [12] Joshua B. Tenenbaum, Vin De Silva, and John C. Langford. A global geometric framework for nonlinear dimensionality reduction. Science, 290(5500):2319–2323, 2000. [13] Jan R. Magnus and Heinz Neudecker. Matrix Differential Calculus with Applications in Statistics and Econometrics. John Wiley & Sons, 2007. [14] Jacek Kierzenka and Lawrence F. Shampine. A BVP solver based on residual control and the Matlab PSE. ACM Transactions on Mathematical Software, 27(3):299–316, 2001. [15] John R. Dormand and P. J. Prince. A family of embedded Runge-Kutta formulae. Journal of Computational and Applied Mathematics, 6:19–26, 1980. [16] P. Thomas Fletcher, Conglin Lu, Stephen M. Pizer, and Sarang Joshi. Principal Geodesic Analysis for the study of Nonlinear Statistics of Shape. IEEE Transactions on Medical Imaging, 23(8):995–1005, 2004. [17] Peter E. Jupp and John T. Kent. Fitting smooth paths to spherical data. Applied Statistics, 36(1):34–46, 1987. [18] Xavier Pennec. Probabilities and statistics on Riemannian manifolds: Basic tools for geometric measurements. In Proceedings of Nonlinear Signal and Image Processing, pages 194–198, 1999. [19] Stefan Sommer, Francois Lauze, Søren Hauberg, and Mads Nielsen. Manifold valued statistics, exact ¸ principal geodesic analysis and the effect of linear approximations. In European Conference on Computer Vision (ECCV), pages 43–56, 2010. [20] Kathleen M. Robinette, Hein Daanen, and Eric Paquet. The CAESAR project: a 3-D surface anthropometry survey. In 3-D Digital Imaging and Modeling, pages 380–386, 1999. [21] Dragomir Anguelov, Praveen Srinivasan, Daphne Koller, Sebastian Thrun, Jim Rodgers, and James Davis. Scape: shape completion and animation of people. ACM Transactions on Graphics, 24(3):408–416, 2005. [22] Oren Freifeld and Michael J. Black. Lie bodies: A manifold representation of 3D human shape. In A. Fitzgibbon et al. (Eds.), editor, European Conference on Computer Vision (ECCV), Part I, LNCS 7572, pages 1–14. Springer-Verlag, oct 2012. [23] Yi Hong, Quannan Li, Jiayan Jiang, and Zhuowen Tu. Learning a mixture of sparse distance metrics for classiﬁcation and dimensionality reduction. In International Conference on Computer Vision (ICCV), pages 906–913, 2011. 9

Reference: text

Summary: the most important sentenses genereted by tfidf model

sentIndex sentText sentNum sentScore

1 de Abstract Multi-metric learning techniques learn local metric tensors in different parts of a feature space. [sent-9, score-0.598]

2 The learned distance measure is, however, non-metric, which has prevented multi-metric learning from generalizing to tasks such as dimensionality reduction and regression in a principled way. [sent-11, score-0.335]

3 Algorithmically, we provide the ﬁrst practical algorithm for computing geodesics according to the learned metrics, as well as algorithms for computing exponential and logarithmic maps on the Riemannian manifold. [sent-14, score-0.714]

4 We illustrate the approach on regression and dimensionality reduction tasks that involve predicting measurements of the human body from shape data. [sent-16, score-0.464]

5 When the Euclidean metric is insufﬁcient, as is the case in many real problems, standard methods break down. [sent-18, score-0.32]

6 This is a key motivation behind metric learning, which strives to learn good distance measures from data. [sent-19, score-0.417]

7 In the most simple scenarios a single metric tensor is learned, but in recent years, several methods have proposed learning multiple metric tensors, such that different distance measures are applied in different parts of the feature space. [sent-20, score-1.007]

8 Speciﬁcally, it is often assumed that straight lines form geodesic curves and that the metric tensor stays constant along these lines. [sent-25, score-1.023]

9 These assumptions are made because it is believed that computing the actual geodesics is intractable, requiring a discretization of the entire feature space [3]. [sent-26, score-0.508]

10 We solve these problems by smoothing the transitions between different metric tensors, which ensures a metric space where geodesics can be computed. [sent-27, score-1.075]

11 In this paper, we consider the scenario where the metric tensor at a given point in feature space is deﬁned as the weighted average of a set of learned metric tensors. [sent-28, score-1.037]

12 Based on the observation that geodesics are 1 (a) Local Metrics & Geodesics (b) Tangent Space Representation (c) First Principal Geodesic Figure 1: Illustration of Principal Geodesic Analysis. [sent-34, score-0.401]

13 (b) Data is mapped to the Euclidean tangent space and the ﬁrst principal component is computed. [sent-36, score-0.391]

14 smooth curves in Riemannian spaces, we derive an algorithm for computing geodesics that only requires a discretization of the geodesic rather than the entire feature space. [sent-38, score-0.817]

15 With this we can map any point back and forth between a Euclidean tangent space and the manifold. [sent-40, score-0.375]

16 This gives us a general strategy for incorporating the learned metric tensors in many Euclidean algorithms: map the data to the tangent of the manifold, perform the Euclidean analysis and map the results back to the manifold. [sent-41, score-0.895]

17 2 Background and Related Work Single-metric learning learns a metric tensor, M, such that distances are measured as dist2 (xi , xj ) = xi − xj 2 M ≡ (xi − xj )T M(xi − xj ) , (2) where M is a symmetric and positive deﬁnite D × D matrix. [sent-49, score-0.519]

18 A more recent approach tries to learn a metric tensor from triplets of data points (xi , xj , xk ), where the metric should obey the constraint that dist(xi , xj ) < dist(xi , xk ). [sent-51, score-1.015]

19 Various relaxed versions of this idea have been suggested such that the metric can be learned by solving a semi-deﬁnite or a quadratic program [1, 2, 4–8]. [sent-53, score-0.388]

20 For many problems, a single global metric tensor is not enough, which motivates learning several local metric tensors. [sent-55, score-0.933]

21 The classic work by Hastie and Tibshirani [9] advocates locally learning metric tensors according to LDA and using these as part of a kNN classiﬁer. [sent-56, score-0.52]

22 In a somewhat similar fashion, Weinberger and Saul [5] cluster the training data and learn a separate metric tensor for each cluster using LMNN. [sent-57, score-0.649]

23 [1, 2], who learn a diagonal metric tensor for every point in the training set, such that distance rankings are preserved. [sent-59, score-0.669]

24 Similarly, Malisiewicz and Efros [6] ﬁnd a diagonal metric tensor for each training point such that the distance to a subset of the training data from the same class is kept small. [sent-60, score-0.644]

25 , MR } has been learned, the distance dist(a, b) is measured according to (2) where “the nearest” metric tensor is used, i. [sent-64, score-0.675]

26 R M(x) = r=1 wr (x) ˜ Mr , where wr (x) = ˜ ˜ j wj (x) 1 0 x − xr 2 r ≤ x − xj M otherwise 2 Mj , ∀j , (3) where x is either a or b depending on the algorithm. [sent-66, score-0.497]

27 The background color is proportional to the trace of the metric tensor, such that light grey corresponds to regions where paths are short (M1 ), and dark grey corresponds to regions they are long (M2 ). [sent-69, score-0.368]

28 Again, background colour is proportional to the trace of the metric tensor; the colour scale is the same is used in (a) and (b). [sent-71, score-0.376]

29 form straight lines, and 2) the metric tensor stays constant along these lines [3]. [sent-73, score-0.736]

30 Assume we are given two metric tensors M1 = 2I and M2 = I positioned at x1 = (2, 2)T and x2 = (4, 4)T respectively. [sent-76, score-0.489]

31 In the same ﬁgure, we also show the assumed straight-line geodesics between selected points in space. [sent-80, score-0.427]

32 As can be seen, two of the lines goes through both regions, such that the assumption of constant metric tensors along the line is violated. [sent-81, score-0.593]

33 Properly computing line lengths according to the local metrics is, however, not enough to ensure that the distance function is metric. [sent-85, score-0.354]

34 2a the straight line does not form a geodesic as a shorter path can be found by circumventing the region with the “expensive” metric tensor M1 as illustrated in Fig. [sent-87, score-0.952]

35 In summary, if we want a metric feature space, we can neither assume that geodesics are straight lines nor that the metric tensor stays constant along such lines. [sent-90, score-1.5]

36 In order to build a model with a metric feature space, we change the weights in (3) to be smooth functions. [sent-95, score-0.419]

37 Formally, they are smooth manifolds endowed with a Riemannian metric [11]: Deﬁnition A Riemannian metric M on a manifold M is a smoothly varying inner product < a, b >x = aT M(x)b in the tangent space Tx M of each point x ∈ M . [sent-100, score-1.219]

38 A geodesic curve is then a length-minimizing curve connecting two given points x and y, i. [sent-107, score-0.391]

39 Given a tangent vector v ∈ Tx M, there exists a unique geodesic cv (t) with initial velocity v at x. [sent-111, score-0.581]

40 The Riemannian exponential map, Expx , maps v to a point on the manifold along the geodesic cv at t = 1. [sent-112, score-0.594]

41 Informally, the exponential and logarithmic maps move points back and forth between the manifold and the tangent space while preserving distances (see Fig. [sent-115, score-0.733]

42 This provides a general strategy for generalizing many Euclidean techniques to Riemannian domains: data points are mapped to the tangent space, where ordinary Euclidean techniques are applied and the results are mapped back to the manifold. [sent-117, score-0.407]

43 We endow C with a metric tensor in every point x, which we deﬁne akin to (3), R M(x) = wr (x)Mr , where wr (x) = r=1 wr (x) ˜ R ˜ j=1 wj (x) , (6) with wr > 0. [sent-120, score-1.346]

44 We assume the metric tensors Mr have already been learned; Sec. [sent-122, score-0.489]

45 From the deﬁnition of a Riemannian metric, we trivially have the following result: Lemma 1 The space C = RD endowed with the metric tensor from (6) is a chart of a Riemannian manifold, iff the weights wr (x) change smoothly with x. [sent-124, score-0.908]

46 Hence, by only considering smooth weight functions wr we get a well-studied geometric structure ˜ on the feature space, which ensures us that it is metric. [sent-125, score-0.318]

47 As can be seen, the metric tensor now changes smoothly, which also makes the geodesics smooth curves (a property we will use when computing the geodesics). [sent-130, score-1.059]

48 It is worth noting that Ramanan and Baker [3] also consider the idea of smoothly averaging the metric tensor. [sent-131, score-0.383]

49 They, however, only evaluate the metric tensor at the test point of their classiﬁer and then assume straight line geodesics with a constant metric tensor. [sent-132, score-1.386]

50 Such assumptions violate the premise of a smoothly changing metric tensor and, again, the distance measure becomes non-metric. [sent-133, score-0.707]

51 Lemma 1 shows that metric learning can be viewed as manifold learning. [sent-134, score-0.483]

52 This gives us the beneﬁt that we can compute geodesics as well as the exponential and logarithmic maps. [sent-136, score-0.53]

53 These provide us with mappings back and forth between the manifold and Euclidean representation of the data, which preserve distances as well as possible. [sent-137, score-0.297]

54 In the next section we will derive a system of ordinary differential equations (ODE’s) that geodesics in C have to satisfy, which provides us with algorithms for computing geodesics as well as exponential and logarithmic maps. [sent-141, score-0.989]

55 From this assumption, we get the following result Theorem 3 For wr (x) = exp − ρ x − xr ˜ 2 ∂vec [M(c)] = ∂c the derivative of the metric tensor from (6) is R ρ R j=1 2 Mr R 2 wj ˜ T r=1 T wj (c − xj ) Mj − (c − xr ) Mr ˜ wr vec [Mr ] ˜ . [sent-152, score-1.218]

56 Hence, to compute a geodesic between x and y, we can solve (8) subject to the constraints c(0) = x and c(1) = y . [sent-156, score-0.287]

57 Our solution avoids discretizing the feature space by discretizing the geodesic curve instead. [sent-160, score-0.481]

58 Once a geodesic c is found, it follows from the deﬁnition of the logarithmic map, Logx (y), that it can be computed as v = Logx (y) = c (0) Length(c) . [sent-163, score-0.376]

59 Given a starting point x on the manifold and a vector v in the tangent space, the exponential map, Expx (v), ﬁnds the unique geodesic starting at x with initial velocity v. [sent-166, score-0.744]

60 As the geodesic must fulﬁll (8), we can compute the exponential map by solving this system of ODE’s with the initial conditions c(0) = x and c (0) = v . [sent-167, score-0.367]

61 1 Generalizing PCA and Regression At this stage, we know that the feature space is Riemannian and we know how to compute geodesics and exponential and logarithmic maps. [sent-171, score-0.607]

62 These generalizations can be summarized as mapping the data to the tangent space at the mean, performing standard Euclidean analysis in the tangent and mapping the results back. [sent-173, score-0.478]

63 The empirical covariance of a set of points is deﬁned as the ordinary Euclidean covariance in the tangent space at the mean value [18]. [sent-176, score-0.31]

64 This approach is called Principal Geodesic Analysis (PGA), and the geodesic curve γvd is called the principal geodesic. [sent-178, score-0.43]

65 Linear regression has been generalized in a similar way [17] by performing regression in the tangent of the mean and mapping the resulting line back to the manifold using the exponential map. [sent-181, score-0.605]

66 The idea of working in the tangent space is both efﬁcient and convenient, but comes with an element of approximation as the logarithmic map is only guarantied to preserve distances to the origin of the tangent and not between all pairs of data points. [sent-182, score-0.662]

67 We sample random points xr and learn a local LDA metric [9] by considering all data points within a radius; this locally pushes the two classes apart. [sent-187, score-0.495]

68 1 Human Body Shape We consider a regression example concerning human body shape analysis. [sent-192, score-0.282]

69 We study 986 female body laser scans from the CAESAR [20] data set; each shape is represented using the leading 35 principal components of the data learned using a SCAPE-like model [21, 22]. [sent-193, score-0.373]

70 We show results for shoulder to wrist distance and shoulder breadth, but results for more measurements are in the supplementary material. [sent-195, score-0.487]

71 To predict the measurements from shape coefﬁcients, we learn local metrics and perform linear regression according to these. [sent-196, score-0.432]

72 As a further experiment, we use PGA to reduce the dimensionality of the shape coefﬁcients according to the local metrics, and measure the quality of the reduction by performing linear regression to predict the measurements. [sent-197, score-0.312]

73 We then cluster the body shapes into equal-sized clusters according to the measurement and learn a LMNN metric for each cluster [5], which we associate with the mean of each class. [sent-201, score-0.551]

74 From this we construct a Riemannian manifold according to (6), 1 Our software implementation for computing geodesics and performing manifold statistics is available at http://ps. [sent-203, score-0.788]

75 (a) Samples from two classes along with illustratively sampled metric tensors from (6). [sent-209, score-0.528]

76 (b) The data represented in the tangent of a manifold constructed from local LDA metrics learned at random positions. [sent-210, score-0.644]

77 (d) Running time (sec) of the geodesic computation as a function of dimensionality. [sent-213, score-0.287]

78 Right: regression error as a function of the dimensionality of the shape space; again the Euclidean metric and the Riemannian metric are compared. [sent-217, score-0.845]

79 compute the mean value on the manifold, map the data to the tangent space at the mean and perform linear regression in the tangent space. [sent-218, score-0.575]

80 4; as can be seen the learned metrics provide principal geodesics, which are more strongly related with the measurements than the Euclidean model. [sent-220, score-0.372]

81 In order to predict the measurements from the body shape, we perform linear regression, both directly in the shape space according to the Euclidean metric and in the tangent space of the manifold corresponding to the learned metrics (using the logarithmic map from (11)). [sent-221, score-1.402]

82 4, showing that regression according to the learned metrics outperforms the Euclidean model. [sent-225, score-0.306]

83 To provide visual insights into the behavior of the learned metrics, we uniformly sample body shape along the ﬁrst principal geodesic (in the range ±7 times the standard deviation) according to the different metrics. [sent-229, score-0.73]

84 As can be seen, the learned metrics pick up intuitive relationships between body shape and the measurements, e. [sent-232, score-0.419]

85 shoulder to wrist distance is related to overall body size, while shoulder breadth is related to body weight. [sent-234, score-0.712]

86 7 Shoulder to wrist distance Shoulder breadth Figure 5: Shapes corresponding to the mean (center) and ±7 times the standard deviations along the principal geodesics (left and right). [sent-235, score-0.748]

87 2 Scalability The human body data set is small enough (986 samples in 35 dimensions) that computing a geodesic only takes a few seconds. [sent-238, score-0.464]

88 We learn one diagonal LMNN metric per class, which we associate with the mean of the class. [sent-241, score-0.345]

89 From this we construct a Riemannian manifold from (6), compute the mean value on the manifold and compute geodesics between the mean and each data point; this is the computationally expensive part of performing PGA. [sent-242, score-0.727]

90 3d plots the average running time (sec) for the computation of geodesics as a function of the dimensionality of the training data. [sent-244, score-0.471]

91 A geodesic can be computed in 100 dimensions in approximately 5 sec. [sent-245, score-0.287]

92 In this experiment, we train a PGA model on 60,000 data points, and test a nearest neighbor classiﬁer in the tangent space as we decrease the dimensionality of the model. [sent-247, score-0.396]

93 The idea of deﬁning the metric tensor at any given point as the weighted average of a ﬁnite set of learned metrics is quite natural from a modeling point of view, which is also validated by the Riemannian structure of the resulting space. [sent-252, score-0.79]

94 This opens both a theoretical and a practical toolbox for analyzing and developing algorithms that use local metric tensors. [sent-253, score-0.361]

95 Speciﬁcally, we show how to use local metric tensors for both regression and dimensionality reduction tasks. [sent-254, score-0.692]

96 For example, Hastie and Tibshirani [9] use local LDA metrics for dimensionality reduction by averaging the local metrics and using the resulting metric as part of a Euclidean PCA, which essentially is a linear approach. [sent-256, score-0.807]

97 [23] who simply compute the principal components according to each metric separately, such that one low dimensional model is learned per metric. [sent-258, score-0.523]

98 Distance metric learning for large margin nearest neighbor classiﬁcation. [sent-286, score-0.39]

99 Manifold valued statistics, exact ¸ principal geodesic analysis and the effect of linear approximations. [sent-342, score-0.391]

100 Learning a mixture of sparse distance metrics for classiﬁcation and dimensionality reduction. [sent-360, score-0.292]

similar papers computed by tfidf model

tfidf for this paper:

wordName wordTfidf (topN-words)

[('geodesics', 0.401), ('riemannian', 0.32), ('metric', 0.32), ('geodesic', 0.287), ('tensor', 0.252), ('tangent', 0.222), ('wr', 0.185), ('tensors', 0.169), ('manifold', 0.163), ('metrics', 0.15), ('dist', 0.142), ('shoulder', 0.131), ('pga', 0.13), ('euclidean', 0.129), ('body', 0.123), ('principal', 0.104), ('logarithmic', 0.089), ('mr', 0.082), ('shape', 0.078), ('ode', 0.074), ('distance', 0.072), ('dimensionality', 0.07), ('learned', 0.068), ('breadth', 0.066), ('freifeld', 0.066), ('wrist', 0.066), ('straight', 0.063), ('smoothly', 0.063), ('vec', 0.058), ('regression', 0.057), ('xr', 0.057), ('hauberg', 0.056), ('logx', 0.056), ('smooth', 0.056), ('distances', 0.055), ('chart', 0.054), ('lmnn', 0.054), ('pca', 0.051), ('measurements', 0.05), ('baker', 0.049), ('lda', 0.047), ('ramanan', 0.047), ('frome', 0.045), ('oren', 0.045), ('feature', 0.043), ('forth', 0.043), ('nearest', 0.043), ('local', 0.041), ('manifolds', 0.041), ('cv', 0.04), ('exponential', 0.04), ('map', 0.04), ('curve', 0.039), ('discretizing', 0.039), ('along', 0.039), ('caesar', 0.037), ('expx', 0.037), ('isomap', 0.037), ('malisiewicz', 0.037), ('metricity', 0.037), ('supplementary', 0.037), ('xj', 0.036), ('back', 0.036), ('reduction', 0.035), ('lines', 0.035), ('geometric', 0.034), ('wj', 0.034), ('knn', 0.034), ('space', 0.034), ('vd', 0.033), ('generalizing', 0.033), ('velocity', 0.032), ('yoram', 0.032), ('mapped', 0.031), ('illustration', 0.031), ('according', 0.031), ('line', 0.03), ('jitendra', 0.03), ('ren', 0.03), ('computing', 0.03), ('classi', 0.029), ('length', 0.029), ('colour', 0.028), ('ordinary', 0.028), ('andrea', 0.027), ('stays', 0.027), ('illustrate', 0.027), ('neighbor', 0.027), ('bingen', 0.026), ('curved', 0.026), ('mpi', 0.026), ('cluster', 0.026), ('points', 0.026), ('spaces', 0.025), ('scatter', 0.025), ('learn', 0.025), ('maps', 0.025), ('regions', 0.024), ('human', 0.024), ('sb', 0.024)]

similar papers list:

simIndex simValue paperId paperTitle

same-paper 1 1.0000002 9 nips-2012-A Geometric take on Metric Learning

Author: Søren Hauberg, Oren Freifeld, Michael J. Black

2 0.39327601 318 nips-2012-Sparse Approximate Manifolds for Differential Geometric MCMC

Author: Ben Calderhead, Mátyás A. Sustik

Abstract: One of the enduring challenges in Markov chain Monte Carlo methodology is the development of proposal mechanisms to make moves distant from the current point, that are accepted with high probability and at low computational cost. The recent introduction of locally adaptive MCMC methods based on the natural underlying Riemannian geometry of such models goes some way to alleviating these problems for certain classes of models for which the metric tensor is analytically tractable, however computational efﬁciency is not assured due to the necessity of potentially high-dimensional matrix operations at each iteration. In this paper we ﬁrstly investigate a sampling-based approach for approximating the metric tensor and suggest a valid MCMC algorithm that extends the applicability of Riemannian Manifold MCMC methods to statistical models that do not admit an analytically computable metric tensor. Secondly, we show how the approximation scheme we consider naturally motivates the use of 1 regularisation to improve estimates and obtain a sparse approximate inverse of the metric, which enables stable and sparse approximations of the local geometry to be made. We demonstrate the application of this algorithm for inferring the parameters of a realistic system of ordinary differential equations using a biologically motivated robust Student-t error model, for which the Expected Fisher Information is analytically intractable. 1

3 0.23600356 265 nips-2012-Parametric Local Metric Learning for Nearest Neighbor Classification

Author: Jun Wang, Alexandros Kalousis, Adam Woznica

Abstract: We study the problem of learning local metrics for nearest neighbor classiﬁcation. Most previous works on local metric learning learn a number of local unrelated metrics. While this ”independence” approach delivers an increased ﬂexibility its downside is the considerable risk of overﬁtting. We present a new parametric local metric learning method in which we learn a smooth metric matrix function over the data manifold. Using an approximation error bound of the metric matrix function we learn local metrics as linear combinations of basis metrics deﬁned on anchor points over different regions of the instance space. We constrain the metric matrix function by imposing on the linear combinations manifold regularization which makes the learned metric matrix function vary smoothly along the geodesics of the data manifold. Our metric learning method has excellent performance both in terms of predictive power and scalability. We experimented with several largescale classiﬁcation problems, tens of thousands of instances, and compared it with several state of the art metric learning methods, both global and local, as well as to SVM with automatic kernel selection, all of which it outperforms in a signiﬁcant manner. 1

4 0.2151642 334 nips-2012-Tensor Decomposition for Fast Parsing with Latent-Variable PCFGs

Author: Michael Collins, Shay B. Cohen

Abstract: We describe an approach to speed-up inference with latent-variable PCFGs, which have been shown to be highly effective for natural language parsing. Our approach is based on a tensor formulation recently introduced for spectral estimation of latent-variable PCFGs coupled with a tensor decomposition algorithm well-known in the multilinear algebra literature. We also describe an error bound for this approximation, which gives guarantees showing that if the underlying tensors are well approximated, then the probability distribution over trees will also be well approximated. Empirical evaluation on real-world natural language parsing data demonstrates a signiﬁcant speed-up at minimal cost for parsing performance. 1

5 0.20395622 25 nips-2012-A new metric on the manifold of kernel matrices with application to matrix geometric means

Author: Suvrit Sra

Abstract: Symmetric positive deﬁnite (spd) matrices pervade numerous scientiﬁc disciplines, including machine learning and optimization. We consider the key task of measuring distances between two spd matrices; a task that is often nontrivial whenever the distance function must respect the non-Euclidean geometry of spd matrices. Typical non-Euclidean distance measures such as the Riemannian metric δR (X, Y ) = log(Y −1/2 XY −1/2 ) F , are computationally demanding and also complicated to use. To allay some of these difﬁculties, we introduce a new metric on spd matrices, which not only respects non-Euclidean geometry but also offers faster computation than δR while being less complicated to use. We support our claims theoretically by listing a set of theorems that relate our metric to δR (X, Y ), and experimentally by studying the nonconvex problem of computing matrix geometric means based on squared distances. 1

6 0.17736621 242 nips-2012-Non-linear Metric Learning

7 0.1606546 307 nips-2012-Semi-Crowdsourced Clustering: Generalizing Crowd Labeling by Robust Distance Metric Learning

8 0.1209268 179 nips-2012-Learning Manifolds with K-Means and K-Flats

9 0.091620184 225 nips-2012-Multi-task Vector Field Learning

10 0.089044362 171 nips-2012-Latent Coincidence Analysis: A Hidden Variable Model for Distance Metric Learning

11 0.087383367 148 nips-2012-Hamming Distance Metric Learning

12 0.078741945 301 nips-2012-Scaled Gradients on Grassmann Manifolds for Matrix Completion

13 0.071837433 78 nips-2012-Compressive Sensing MRI with Wavelet Tree Sparsity

14 0.069596745 157 nips-2012-Identification of Recurrent Patterns in the Activation of Brain Networks

15 0.069039866 184 nips-2012-Learning Probability Measures with respect to Optimal Transport Metrics

16 0.067291543 187 nips-2012-Learning curves for multi-task Gaussian process regression

17 0.061612319 145 nips-2012-Gradient Weights help Nonparametric Regressors

18 0.060271256 200 nips-2012-Local Supervised Learning through Space Partitioning

19 0.059179366 360 nips-2012-Visual Recognition using Embedded Feature Selection for Curvature Self-Similarity

20 0.052804664 146 nips-2012-Graphical Gaussian Vector for Image Categorization

similar papers computed by lsi model

lsi for this paper:

topicId topicWeight

[(0, 0.175), (1, 0.061), (2, -0.032), (3, -0.105), (4, 0.045), (5, -0.003), (6, -0.005), (7, 0.082), (8, 0.1), (9, -0.049), (10, 0.039), (11, -0.271), (12, 0.071), (13, -0.11), (14, -0.159), (15, 0.154), (16, -0.045), (17, 0.157), (18, 0.375), (19, 0.147), (20, 0.141), (21, -0.095), (22, -0.005), (23, 0.026), (24, 0.085), (25, 0.0), (26, 0.001), (27, -0.021), (28, -0.027), (29, -0.016), (30, -0.072), (31, -0.016), (32, -0.022), (33, 0.06), (34, -0.004), (35, -0.013), (36, 0.107), (37, 0.02), (38, -0.039), (39, -0.138), (40, 0.051), (41, 0.033), (42, -0.001), (43, -0.018), (44, -0.051), (45, -0.064), (46, -0.057), (47, -0.022), (48, 0.015), (49, -0.034)]

similar papers list:

simIndex simValue paperId paperTitle

same-paper 1 0.9597944 9 nips-2012-A Geometric take on Metric Learning

Author: Søren Hauberg, Oren Freifeld, Michael J. Black

2 0.82496643 318 nips-2012-Sparse Approximate Manifolds for Differential Geometric MCMC

Author: Ben Calderhead, Mátyás A. Sustik

3 0.74911416 25 nips-2012-A new metric on the manifold of kernel matrices with application to matrix geometric means

Author: Suvrit Sra

4 0.74509966 265 nips-2012-Parametric Local Metric Learning for Nearest Neighbor Classification

Author: Jun Wang, Alexandros Kalousis, Adam Woznica

5 0.5978809 242 nips-2012-Non-linear Metric Learning

Author: Dor Kedem, Stephen Tyree, Fei Sha, Gert R. Lanckriet, Kilian Q. Weinberger

Abstract: In this paper, we introduce two novel metric learning algorithms, χ2 -LMNN and GB-LMNN, which are explicitly designed to be non-linear and easy-to-use. The two approaches achieve this goal in fundamentally different ways: χ2 -LMNN inherits the computational beneﬁts of a linear mapping from linear metric learning, but uses a non-linear χ2 -distance to explicitly capture similarities within histogram data sets; GB-LMNN applies gradient-boosting to learn non-linear mappings directly in function space and takes advantage of this approach’s robustness, speed, parallelizability and insensitivity towards the single additional hyperparameter. On various benchmark data sets, we demonstrate these methods not only match the current state-of-the-art in terms of kNN classiﬁcation error, but in the case of χ2 -LMNN, obtain best results in 19 out of 20 learning settings. 1

6 0.55510908 307 nips-2012-Semi-Crowdsourced Clustering: Generalizing Crowd Labeling by Robust Distance Metric Learning

7 0.52090865 334 nips-2012-Tensor Decomposition for Fast Parsing with Latent-Variable PCFGs

8 0.51198089 225 nips-2012-Multi-task Vector Field Learning

9 0.45645872 171 nips-2012-Latent Coincidence Analysis: A Hidden Variable Model for Distance Metric Learning

10 0.42648941 179 nips-2012-Learning Manifolds with K-Means and K-Flats

11 0.39435145 157 nips-2012-Identification of Recurrent Patterns in the Activation of Brain Networks

12 0.38547719 146 nips-2012-Graphical Gaussian Vector for Image Categorization

13 0.37563866 22 nips-2012-A latent factor model for highly multi-relational data

14 0.37463984 301 nips-2012-Scaled Gradients on Grassmann Manifolds for Matrix Completion

15 0.35013488 148 nips-2012-Hamming Distance Metric Learning

16 0.34210977 145 nips-2012-Gradient Weights help Nonparametric Regressors

17 0.32536808 54 nips-2012-Bayesian Probabilistic Co-Subspace Addition

18 0.32353836 184 nips-2012-Learning Probability Measures with respect to Optimal Transport Metrics

19 0.32072198 338 nips-2012-The Perturbed Variation

20 0.30855513 176 nips-2012-Learning Image Descriptors with the Boosting-Trick

similar papers computed by lda model

lda for this paper:

topicId topicWeight

[(0, 0.031), (21, 0.027), (38, 0.104), (42, 0.076), (54, 0.028), (55, 0.01), (72, 0.295), (74, 0.074), (76, 0.154), (80, 0.066), (92, 0.033)]

similar papers list:

simIndex simValue paperId paperTitle

same-paper 1 0.7819671 9 nips-2012-A Geometric take on Metric Learning

Author: Søren Hauberg, Oren Freifeld, Michael J. Black

2 0.76628399 107 nips-2012-Effective Split-Merge Monte Carlo Methods for Nonparametric Models of Sequential Data

Author: Michael C. Hughes, Erik B. Sudderth, Emily B. Fox

Abstract: Applications of Bayesian nonparametric methods require learning and inference algorithms which efﬁciently explore models of unbounded complexity. We develop new Markov chain Monte Carlo methods for the beta process hidden Markov model (BP-HMM), enabling discovery of shared activity patterns in large video and motion capture databases. By introducing split-merge moves based on sequential allocation, we allow large global changes in the shared feature structure. We also develop data-driven reversible jump moves which more reliably discover rare or unique behaviors. Our proposals apply to any choice of conjugate likelihood for observed data, and we show success with multinomial, Gaussian, and autoregressive emission models. Together, these innovations allow tractable analysis of hundreds of time series, where previous inference required clever initialization and lengthy burn-in periods for just six sequences. 1

3 0.75043607 101 nips-2012-Discriminatively Trained Sparse Code Gradients for Contour Detection

Author: Ren Xiaofeng, Liefeng Bo

Abstract: Finding contours in natural images is a fundamental problem that serves as the basis of many tasks such as image segmentation and object recognition. At the core of contour detection technologies are a set of hand-designed gradient features, used by most approaches including the state-of-the-art Global Pb (gPb) operator. In this work, we show that contour detection accuracy can be significantly improved by computing Sparse Code Gradients (SCG), which measure contrast using patch representations automatically learned through sparse coding. We use K-SVD for dictionary learning and Orthogonal Matching Pursuit for computing sparse codes on oriented local neighborhoods, and apply multi-scale pooling and power transforms before classifying them with linear SVMs. By extracting rich representations from pixels and avoiding collapsing them prematurely, Sparse Code Gradients effectively learn how to measure local contrasts and ﬁnd contours. We improve the F-measure metric on the BSDS500 benchmark to 0.74 (up from 0.71 of gPb contours). Moreover, our learning approach can easily adapt to novel sensor data such as Kinect-style RGB-D cameras: Sparse Code Gradients on depth maps and surface normals lead to promising contour detection using depth and depth+color, as veriﬁed on the NYU Depth Dataset. 1

4 0.74579948 250 nips-2012-On-line Reinforcement Learning Using Incremental Kernel-Based Stochastic Factorization

Author: Doina Precup, Joelle Pineau, Andre S. Barreto

Abstract: Kernel-based stochastic factorization (KBSF) is an algorithm for solving reinforcement learning tasks with continuous state spaces which builds a Markov decision process (MDP) based on a set of sample transitions. What sets KBSF apart from other kernel-based approaches is the fact that the size of its MDP is independent of the number of transitions, which makes it possible to control the trade-off between the quality of the resulting approximation and the associated computational cost. However, KBSF’s memory usage grows linearly with the number of transitions, precluding its application in scenarios where a large amount of data must be processed. In this paper we show that it is possible to construct KBSF’s MDP in a fully incremental way, thus freeing the space complexity of this algorithm from its dependence on the number of sample transitions. The incremental version of KBSF is able to process an arbitrary amount of data, which results in a model-based reinforcement learning algorithm that can be used to solve continuous MDPs in both off-line and on-line regimes. We present theoretical results showing that KBSF can approximate the value function that would be computed by conventional kernel-based learning with arbitrary precision. We empirically demonstrate the effectiveness of the proposed algorithm in the challenging threepole balancing task, in which the ability to process a large number of transitions is crucial for success. 1

5 0.67666805 258 nips-2012-Online L1-Dictionary Learning with Application to Novel Document Detection

Author: Shiva P. Kasiviswanathan, Huahua Wang, Arindam Banerjee, Prem Melville

Abstract: Given their pervasive use, social media, such as Twitter, have become a leading source of breaking news. A key task in the automated identiﬁcation of such news is the detection of novel documents from a voluminous stream of text documents in a scalable manner. Motivated by this challenge, we introduce the problem of online 1 -dictionary learning where unlike traditional dictionary learning, which uses squared loss, the 1 -penalty is used for measuring the reconstruction error. We present an efﬁcient online algorithm for this problem based on alternating directions method of multipliers, and establish a sublinear regret bound for this algorithm. Empirical results on news-stream and Twitter data, shows that this online 1 -dictionary learning algorithm for novel document detection gives more than an order of magnitude speedup over the previously known batch algorithm, without any signiﬁcant loss in quality of results. 1

6 0.63947952 178 nips-2012-Learning Label Trees for Probabilistic Modelling of Implicit Feedback

7 0.63468301 318 nips-2012-Sparse Approximate Manifolds for Differential Geometric MCMC

8 0.60683537 54 nips-2012-Bayesian Probabilistic Co-Subspace Addition

9 0.59253645 242 nips-2012-Non-linear Metric Learning

10 0.59178752 209 nips-2012-Max-Margin Structured Output Regression for Spatio-Temporal Action Localization

11 0.59015816 289 nips-2012-Recognizing Activities by Attribute Dynamics

12 0.58607727 219 nips-2012-Modelling Reciprocating Relationships with Hawkes Processes

13 0.58591652 68 nips-2012-Clustering Aggregation as Maximum-Weight Independent Set

14 0.58522075 188 nips-2012-Learning from Distributions via Support Measure Machines

15 0.58493793 316 nips-2012-Small-Variance Asymptotics for Exponential Family Dirichlet Process Mixture Models

16 0.58491993 38 nips-2012-Algorithms for Learning Markov Field Policies

17 0.58476835 210 nips-2012-Memorability of Image Regions

18 0.58461094 162 nips-2012-Inverse Reinforcement Learning through Structured Classification

19 0.58302891 90 nips-2012-Deep Learning of Invariant Features via Simulated Fixations in Video

20 0.58281672 176 nips-2012-Learning Image Descriptors with the Boosting-Trick