hunch_net hunch_net-2008 hunch_net-2008-284 knowledge-graph by maker-knowledge-mining
Source: html
Introduction: David Pennock notes the impressive set of datasets at datawrangling .
sentIndex sentText sentNum sentScore
1 David Pennock notes the impressive set of datasets at datawrangling . [sent-1, score-1.719]
wordName wordTfidf (topN-words)
[('impressive', 0.549), ('notes', 0.537), ('david', 0.435), ('datasets', 0.419), ('set', 0.214)]
simIndex simValue blogId blogTitle
same-blog 1 1.0 284 hunch net-2008-01-18-Datasets
Introduction: David Pennock notes the impressive set of datasets at datawrangling .
2 0.17839091 486 hunch net-2013-07-10-Thoughts on Artificial Intelligence
Introduction: David McAllester starts a blog .
3 0.14785241 19 hunch net-2005-02-14-Clever Methods of Overfitting
Introduction: “Overfitting” is traditionally defined as training some flexible representation so that it memorizes the data but fails to predict well in the future. For this post, I will define overfitting more generally as over-representing the performance of systems. There are two styles of general overfitting: overrepresenting performance on particular datasets and (implicitly) overrepresenting performance of a method on future datasets. We should all be aware of these methods, avoid them where possible, and take them into account otherwise. I have used “reproblem” and “old datasets”, and may have participated in “overfitting by review”—some of these are very difficult to avoid. Name Method Explanation Remedy Traditional overfitting Train a complex predictor on too-few examples. Hold out pristine examples for testing. Use a simpler predictor. Get more training examples. Integrate over many predictors. Reject papers which do this. Parameter twe
4 0.12191123 200 hunch net-2006-08-03-AOL’s data drop
Introduction: AOL has released several large search engine related datasets. This looks like a pretty impressive data release, and it is a big opportunity for people everywhere to worry about search engine related learning problems, if they want.
5 0.11149674 251 hunch net-2007-06-24-Interesting Papers at ICML 2007
Introduction: Here are a few of the papers I enjoyed at ICML. Steffen Bickel , Michael Brüeckner, Tobias Scheffer , Discriminative Learning for Differing Training and Test Distributions There is a nice trick in this paper: they predict the probability that an unlabeled sample is in the training set vs. the test set, and then use this prediction to importance weight labeled samples in the training set. This paper uses a specific parametric model, but the approach is easily generalized. Steve Hanneke A Bound on the Label Complexity of Agnostic Active Learning This paper bounds the number of labels required by the A 2 algorithm for active learning in the agnostic case. Last year we figured out agnostic active learning was possible. This year, it’s quantified. Hopefull soon, it will be practical. Sylvian Gelly , David Silver Combining Online and Offline Knowledge in UCT . This paper is about techniques for improving MoGo with various sorts of learning. MoGo has a fair
6 0.071852222 193 hunch net-2006-07-09-The Stock Prediction Machine Learning Problem
7 0.070244342 422 hunch net-2011-01-16-2011 Summer Conference Deadline Season
8 0.069192402 129 hunch net-2005-11-07-Prediction Competitions
9 0.069126964 300 hunch net-2008-04-30-Concerns about the Large Scale Learning Challenge
10 0.066005543 255 hunch net-2007-07-13-The View From China
11 0.061860919 131 hunch net-2005-11-16-The Everything Ensemble Edge
12 0.058820173 393 hunch net-2010-04-14-MLcomp: a website for objectively comparing ML algorithms
13 0.056726277 188 hunch net-2006-06-30-ICML papers
14 0.05618066 375 hunch net-2009-10-26-NIPS workshops
15 0.055017285 406 hunch net-2010-08-22-KDD 2010
16 0.054667961 454 hunch net-2012-01-30-ICML Posters and Scope
17 0.051812291 410 hunch net-2010-09-17-New York Area Machine Learning Events
18 0.051544018 211 hunch net-2006-10-02-$1M Netflix prediction contest
19 0.051514853 310 hunch net-2008-07-15-Interesting papers at COLT (and a bit of UAI & workshops)
20 0.05084743 400 hunch net-2010-06-13-The Good News on Exploration and Learning
topicId topicWeight
[(0, 0.047), (1, 0.017), (2, -0.021), (3, -0.01), (4, 0.022), (5, -0.015), (6, -0.05), (7, -0.037), (8, 0.021), (9, -0.011), (10, -0.029), (11, 0.129), (12, 0.012), (13, -0.007), (14, -0.013), (15, 0.004), (16, 0.036), (17, 0.026), (18, -0.019), (19, 0.027), (20, 0.033), (21, -0.062), (22, -0.033), (23, -0.007), (24, 0.011), (25, 0.021), (26, 0.009), (27, -0.077), (28, -0.058), (29, -0.022), (30, -0.001), (31, 0.081), (32, 0.111), (33, 0.005), (34, 0.126), (35, 0.014), (36, -0.04), (37, -0.019), (38, 0.07), (39, 0.108), (40, 0.166), (41, -0.004), (42, -0.069), (43, -0.006), (44, 0.005), (45, -0.064), (46, -0.146), (47, -0.049), (48, -0.099), (49, 0.047)]
simIndex simValue blogId blogTitle
same-blog 1 0.99163461 284 hunch net-2008-01-18-Datasets
Introduction: David Pennock notes the impressive set of datasets at datawrangling .
2 0.63050938 19 hunch net-2005-02-14-Clever Methods of Overfitting
Introduction: “Overfitting” is traditionally defined as training some flexible representation so that it memorizes the data but fails to predict well in the future. For this post, I will define overfitting more generally as over-representing the performance of systems. There are two styles of general overfitting: overrepresenting performance on particular datasets and (implicitly) overrepresenting performance of a method on future datasets. We should all be aware of these methods, avoid them where possible, and take them into account otherwise. I have used “reproblem” and “old datasets”, and may have participated in “overfitting by review”—some of these are very difficult to avoid. Name Method Explanation Remedy Traditional overfitting Train a complex predictor on too-few examples. Hold out pristine examples for testing. Use a simpler predictor. Get more training examples. Integrate over many predictors. Reject papers which do this. Parameter twe
3 0.44778663 131 hunch net-2005-11-16-The Everything Ensemble Edge
Introduction: Rich Caruana , Alexandru Niculescu , Geoff Crew, and Alex Ksikes have done a lot of empirical testing which shows that using all methods to make a prediction is more powerful than using any single method. This is in rough agreement with the Bayesian way of solving problems, but based upon a different (essentially empirical) motivation. A rough summary is: Take all of {decision trees, boosted decision trees, bagged decision trees, boosted decision stumps, K nearest neighbors, neural networks, SVM} with all reasonable parameter settings. Run the methods on each problem of 8 problems with a large test set, calibrating margins using either sigmoid fitting or isotonic regression . For each loss of {accuracy, area under the ROC curve, cross entropy, squared error, etc…} evaluate the average performance of the method. A series of conclusions can be drawn from the observations. ( Calibrated ) boosted decision trees appear to perform best, in general although support v
4 0.4183442 362 hunch net-2009-06-26-Netflix nearly done
Introduction: A $1M qualifying result was achieved on the public Netflix test set by a 3-way ensemble team . This is just in time for Yehuda ‘s presentation at KDD , which I’m sure will be one of the best attended ever. This isn’t quite over—there are a few days for another super-conglomerate team to come together and there is some small chance that the performance is nonrepresentative of the final test set, but I expect not. Regardless of the final outcome, the biggest lesson for ML from the Netflix contest has been the formidable performance edge of ensemble methods.
5 0.37951005 211 hunch net-2006-10-02-$1M Netflix prediction contest
Introduction: Netflix is running a contest to improve recommender prediction systems. A 10% improvement over their current system yields a $1M prize. Failing that, the best smaller improvement yields a smaller $50K prize. This contest looks quite real, and the $50K prize money is almost certainly achievable with a bit of thought. The contest also comes with a dataset which is apparently 2 orders of magnitude larger than any other public recommendation system datasets.
6 0.37569514 18 hunch net-2005-02-12-ROC vs. Accuracy vs. AROC
7 0.37141412 486 hunch net-2013-07-10-Thoughts on Artificial Intelligence
8 0.37067014 129 hunch net-2005-11-07-Prediction Competitions
9 0.35985252 251 hunch net-2007-06-24-Interesting Papers at ICML 2007
10 0.35390645 56 hunch net-2005-04-14-Families of Learning Theory Statements
11 0.35197836 239 hunch net-2007-04-18-$50K Spock Challenge
12 0.34989786 300 hunch net-2008-04-30-Concerns about the Large Scale Learning Challenge
13 0.33495378 393 hunch net-2010-04-14-MLcomp: a website for objectively comparing ML algorithms
14 0.33208561 26 hunch net-2005-02-21-Problem: Cross Validation
15 0.33186367 399 hunch net-2010-05-20-Google Predict
16 0.32096979 349 hunch net-2009-04-21-Interesting Presentations at Snowbird
17 0.30086595 364 hunch net-2009-07-11-Interesting papers at KDD
18 0.30046475 177 hunch net-2006-05-05-An ICML reject
19 0.27420723 406 hunch net-2010-08-22-KDD 2010
20 0.27038646 418 hunch net-2010-12-02-Traffic Prediction Problem
topicId topicWeight
[(90, 0.686)]
simIndex simValue blogId blogTitle
1 0.92294157 323 hunch net-2008-11-04-Rise of the Machines
Introduction: On the enduring topic of how people deal with intelligent machines , we have this important election bulletin .
2 0.89859974 340 hunch net-2009-01-28-Nielsen’s talk
Introduction: I wanted to point to Michael Nielsen’s talk about blogging science, which I found interesting.
same-blog 3 0.78501713 284 hunch net-2008-01-18-Datasets
Introduction: David Pennock notes the impressive set of datasets at datawrangling .
4 0.7049492 32 hunch net-2005-02-27-Antilearning: When proximity goes bad
Introduction: Joel Predd mentioned “ Antilearning ” by Adam Kowalczyk , which is interesting from a foundational intuitions viewpoint. There is a pervasive intuition that “nearby things tend to have the same label”. This intuition is instantiated in SVMs, nearest neighbor classifiers, decision trees, and neural networks. It turns out there are natural problems where this intuition is opposite of the truth. One natural situation where this occurs is in competition. For example, when Intel fails to meet its earnings estimate, is this evidence that AMD is doing badly also? Or evidence that AMD is doing well? This violation of the proximity intuition means that when the number of examples is few, negating a classifier which attempts to exploit proximity can provide predictive power (thus, the term “antilearning”).
5 0.68387479 239 hunch net-2007-04-18-$50K Spock Challenge
Introduction: Apparently, the company Spock is setting up a $50k entity resolution challenge . $50k is much less than the Netflix challenge, but it’s effectively the same as Netflix until someone reaches 10% . It’s also nice that the Spock challenge has a short duration. The (visible) test set is of size 25k and the training set has size 75k.
6 0.52578551 139 hunch net-2005-12-11-More NIPS Papers
7 0.48276007 144 hunch net-2005-12-28-Yet more nips thoughts
8 0.28467268 333 hunch net-2008-12-27-Adversarial Academia
9 0.037052784 422 hunch net-2011-01-16-2011 Summer Conference Deadline Season
10 0.032866511 134 hunch net-2005-12-01-The Webscience Future
11 0.024599608 95 hunch net-2005-07-14-What Learning Theory might do
12 0.0 1 hunch net-2005-01-19-Why I decided to run a weblog.
13 0.0 2 hunch net-2005-01-24-Holy grails of machine learning?
14 0.0 3 hunch net-2005-01-24-The Humanloop Spectrum of Machine Learning
15 0.0 4 hunch net-2005-01-26-Summer Schools
16 0.0 5 hunch net-2005-01-26-Watchword: Probability
17 0.0 6 hunch net-2005-01-27-Learning Complete Problems
18 0.0 7 hunch net-2005-01-31-Watchword: Assumption
19 0.0 8 hunch net-2005-02-01-NIPS: Online Bayes
20 0.0 9 hunch net-2005-02-01-Watchword: Loss