nips nips2005 nips2005-118 knowledge-graph by maker-knowledge-mining

118 nips-2005-Learning in Silicon: Timing is Everything

Source: pdf

Author: John V. Arthur, Kwabena Boahen

Abstract: We describe a neuromorphic chip that uses binary synapses with spike timing-dependent plasticity (STDP) to learn stimulated patterns of activity and to compensate for variability in excitability. Speciﬁcally, STDP preferentially potentiates (turns on) synapses that project from excitable neurons, which spike early, to lethargic neurons, which spike late. The additional excitatory synaptic current makes lethargic neurons spike earlier, thereby causing neurons that belong to the same pattern to spike in synchrony. Once learned, an entire pattern can be recalled by stimulating a subset. 1 Variability in Neural Systems Evidence suggests precise spike timing is important in neural coding, speciﬁcally, in the hippocampus. The hippocampus uses timing in the spike activity of place cells (in addition to rate) to encode location in space [1]. Place cells employ a phase code: the timing at which a neuron spikes relative to the phase of the inhibitory theta rhythm (5-12Hz) conveys information. As an animal approaches a place cell’s preferred location, the place cell not only increases its spike rate, but also spikes at earlier phases in the theta cycle. To implement a phase code, the theta rhythm is thought to prevent spiking until the input synaptic current exceeds the sum of the neuron threshold and the decreasing inhibition on the downward phase of the cycle [2]. However, even with identical inputs and common theta inhibition, neurons do not spike in synchrony. Variability in excitability spreads the activity in phase. Lethargic neurons (such as those with high thresholds) spike late in the theta cycle, since their input exceeds the sum of the neuron threshold and theta inhibition only after the theta inhibition has had time to decrease. Conversely, excitable neurons (such as those with low thresholds) spike early in the theta cycle. Consequently, variability in excitability translates into variability in timing. We hypothesize that the hippocampus achieves its precise spike timing (about 10ms) through plasticity enhanced phase-coding (PEP). The source of hippocampal timing precision in the presence of variability (and noise) remains unexplained. Synaptic plasticity can compensate for variability in excitability if it increases excitatory synaptic input to neurons in inverse proportion to their excitabilities. Recasting this in a phase-coding framework, we desire a learning rule that increases excitatory synaptic input to neurons directly related to their phases. Neurons that lag require additional synaptic input, whereas neurons that lead 120µm 190µm A B Figure 1: STDP Chip. A The chip has a 16-by-16 array of microcircuits; one microcircuit includes four principal neurons, each with 21 STDP circuits. B The STDP Chip is embedded in a circuit board including DACs, a CPLD, a RAM chip, and a USB chip, which communicates with a PC. require none. The spike timing-dependent plasticity (STDP) observed in the hippocampus satisﬁes this requirement [3]. It requires repeated pre-before-post spike pairings (within a time window) to potentiate and repeated post-before-pre pairings to depress a synapse. Here we validate our hypothesis with a model implemented in silicon, where variability is as ubiquitous as it is in biology [4]. Section 2 presents our silicon system, including the STDP Chip. Section 3 describes and characterizes the STDP circuit. Section 4 demonstrates that PEP compensates for variability and provides evidence that STDP is the compensation mechanism. Section 5 explores a desirable consequence of PEP: unconventional associative pattern recall. Section 6 discusses the implications of the PEP model, including its beneﬁts and applications in the engineering of neuromorphic systems and in the study of neurobiology. 2 Silicon System We have designed, submitted, and tested a silicon implementation of PEP. The STDP Chip was fabricated through MOSIS in a 1P5M 0.25µm CMOS process, with just under 750,000 transistors in just over 10mm2 of area. It has a 32 by 32 array of excitatory principal neurons commingled with a 16 by 16 array of inhibitory interneurons that are not used here (Figure 1A). Each principal neuron has 21 STDP synapses. The address-event representation (AER) [5] is used to transmit spikes off chip and to receive afferent and recurrent spike input. To conﬁgure the STDP Chip as a recurrent network, we embedded it in a circuit board (Figure 1B). The board has ﬁve primary components: a CPLD (complex programmable logic device), the STDP Chip, a RAM chip, a USB interface chip, and DACs (digital-to-analog converters). The central component in the system is the CPLD. The CPLD handles AER trafﬁc, mediates communication between devices, and implements recurrent connections by accessing a lookup table, stored in the RAM chip. The USB interface chip provides a bidirectional link with a PC. The DACs control the analog biases in the system, including the leak current, which the PC varies in real-time to create the global inhibitory theta rhythm. The principal neuron consists of a refractory period and calcium-dependent potassium circuit (RCK), a synapse circuit, and a soma circuit (Figure 2A). RCK and the synapse are ISOMA Soma Synapse STDP Presyn. Spike PE LPF A Presyn. Spike Raster AH 0 0.1 Spike probability RCK Postsyn. Spike B 0.05 0.1 0.05 0.1 0.08 0.06 0.04 0.02 0 0 Time(s) Figure 2: Principal neuron. A A simpliﬁed schematic is shown, including: the synapse, refractory and calcium-dependent potassium channel (RCK), soma, and axon-hillock (AH) circuits, plus their constituent elements, the pulse extender (PE) and the low-pass ﬁlter (LPF). B Spikes (dots) from 81 principal neurons are temporally dispersed, when excited by poisson-like inputs (58Hz) and inhibited by the common 8.3Hz theta rhythm (solid line). The histogram includes spikes from ﬁve theta cycles. composed of two reusable blocks: the low-pass ﬁlter (LPF) and the pulse extender (PE). The soma is a modiﬁed version of the LPF, which receives additional input from an axonhillock circuit (AH). RCK is inhibitory to the neuron. It consists of a PE, which models calcium inﬂux during a spike, and a LPF, which models calcium buffering. When AH ﬁres a spike, a packet of charge is dumped onto a capacitor in the PE. The PE’s output activates until the charge decays away, which takes a few milliseconds. Also, while the PE is active, charge accumulates on the LPF’s capacitor, lowering the LPF’s output voltage. Once the PE deactivates, this charge leaks away as well, but this takes tens of milliseconds because the leak is smaller. The PE’s and the LPF’s inhibitory effects on the soma are both described below in terms of the sum (ISHUNT ) of the currents their output voltages produce in pMOS transistors whose sources are at Vdd (see Figure 2A). Note that, in the absence of spikes, these currents decay exponentially, with a time-constant determined by their respective leaks. The synapse circuit is excitatory to the neuron. It is composed of a PE, which represents the neurotransmitter released into the synaptic cleft, and a LPF, which represents the bound neurotransmitter. The synapse circuit is similar to RCK in structure but differs in function: It is activated not by the principal neuron itself but by the STDP circuits (or directly by afferent spikes that bypass these circuits, i.e., ﬁxed synapses). The synapse’s effect on the soma is also described below in terms of the current (ISYN ) its output voltage produces in a pMOS transistor whose source is at Vdd. The soma circuit is a leaky integrator. It receives excitation from the synapse circuit and shunting inhibition from RCK and has a leak current as well. Its temporal behavior is described by: τ dISOMA ISYN I0 + ISOMA = dt ISHUNT where ISOMA is the current the capacitor’s voltage produces in a pMOS transistor whose source is at Vdd (see Figure 2A). ISHUNT is the sum of the leak, refractory, and calciumdependent potassium currents. These currents also determine the time constant: τ = C Ut κISHUNT , where I0 and κ are transistor parameters and Ut is the thermal voltage. STDP circuit ~LTP SRAM Presynaptic spike A ~LTD Inverse number of pairings Integrator Decay Postsynaptic spike Potentiation 0.1 0.05 0 0.05 0.1 Depression -80 -40 0 Presynaptic spike Postsynaptic spike 40 Spike timing: t pre - t post (ms) 80 B Figure 3: STDP circuit design and characterization. A The circuit is composed of three subcircuits: decay, integrator, and SRAM. B The circuit potentiates when the presynaptic spike precedes the postsynaptic spike and depresses when the postsynaptic spike precedes the presynaptic spike. The soma circuit is connected to an AH, the locus of spike generation. The AH consists of model voltage-dependent sodium and potassium channel populations (modiﬁed from [6] by Kai Hynna). It initiates the AER signaling process required to send a spike off chip. To characterize principal neuron variability, we excited 81 neurons with poisson-like 58Hz spike trains (Figure 2B). We made these spike trains poisson-like by starting with a regular 200Hz spike train and dropping spikes randomly, with probability of 0.71. Thus spikes were delivered to neurons that won the coin toss in synchrony every 5ms. However, neurons did not lock onto the input synchrony due to ﬁltering by the synaptic time constant (see Figure 2B). They also received a common inhibitory input at the theta frequency (8.3Hz), via their leak current. Each neuron was prevented from ﬁring more than one spike in a theta cycle by its model calcium-dependent potassium channel population. The principal neurons’ spike times were variable. To quantify the spike variability, we used timing precision, which we deﬁne as twice the standard deviation of spike times accumulated from ﬁve theta cycles. With an input rate of 58Hz the timing precision was 34ms. 3 STDP Circuit The STDP circuit (related to [7]-[8]), for which the STDP Chip is named, is the most abundant, with 21,504 copies on the chip. This circuit is built from three subcircuits: decay, integrator, and SRAM (Figure 3A). The decay and integrator are used to implement potentiation, and depression, in a symmetric fashion. The SRAM holds the current binary state of the synapse, either potentiated or depressed. For potentiation, the decay remembers the last presynaptic spike. Its capacitor is charged when that spike occurs and discharges linearly thereafter. A postsynaptic spike samples the charge remaining on the capacitor, passes it through an exponential function, and dumps the resultant charge into the integrator. This charge decays linearly thereafter. At the time of the postsynaptic spike, the SRAM, a cross-coupled inverter pair, reads the voltage on the integrator’s capacitor. If it exceeds a threshold, the SRAM switches state from depressed to potentiated (∼LTD goes high and ∼LTP goes low). The depression side of the STDP circuit is exactly symmetric, except that it responds to postsynaptic activation followed by presynaptic activation and switches the SRAM’s state from potentiated to depressed (∼LTP goes high and ∼LTD goes low). When the SRAM is in the potentiated state, the presynaptic 50 After STDP 83 92 100 Timing precision(ms) Before STDP 75 B Before STDP After STDP 40 30 20 10 0 50 60 70 80 90 Input rate(Hz) 100 50 58 67 text A 0.2 0.4 Time(s) 0.6 0.2 0.4 Time(s) 0.6 C Figure 4: Plasticity enhanced phase-coding. A Spike rasters of 81 neurons (9 by 9 cluster) display synchrony over a two-fold range of input rates after STDP. B The degree of enhancement is quantiﬁed by timing precision. C Each neuron (center box) sends synapses to (dark gray) and receives synapses from (light gray) twenty-one randomly chosen neighbors up to ﬁve nodes away (black indicates both connections). spike activates the principal neuron’s synapse; otherwise the spike has no effect. We characterized the STDP circuit by activating a plastic synapse and a ﬁxed synapse– which elicits a spike at different relative times. We repeated this pairing at 16Hz. We counted the number of pairings required to potentiate (or depress) the synapse. Based on this count, we calculated the efﬁcacy of each pairing as the inverse number of pairings required (Figure 3B). For example, if twenty pairings were required to potentiate the synapse, the efﬁcacy of that pre-before-post time-interval was one twentieth. The efﬁcacy of both potentiation and depression are ﬁt by exponentials with time constants of 11.4ms and 94.9ms, respectively. This behavior is similar to that observed in the hippocampus: potentiation has a shorter time constant and higher maximum efﬁcacy than depression [3]. 4 Recurrent Network We carried out an experiment designed to test the STDP circuit’s ability to compensate for variability in spike timing through PEP. Each neuron received recurrent connections from 21 randomly selected neurons within an 11 by 11 neighborhood centered on itself (see Figure 4C). Conversely, it made recurrent connections to randomly chosen neurons within the same neighborhood. These connections were mediated by STDP circuits, initialized to the depressed state. We chose a 9 by 9 cluster of neurons and delivered spikes at a mean rate of 50 to 100Hz to each one (dropping spikes with a probability of 0.75 to 0.5 from a regular 200Hz train) and provided common theta inhibition as before. We compared the variability in spike timing after ﬁve seconds of learning with the initial distribution. Phase coding was enhanced after STDP (Figure 4A). Before STDP, spike timing among neurons was highly variable (except for the very highest input rate). After STDP, variability was virtually eliminated (except for the very lowest input rate). Initially, the variability, characterized by timing precision, was inversely related to the input rate, decreasing from 34 to 13ms. After ﬁve seconds of STDP, variability decreased and was largely independent of input rate, remaining below 11ms. Potentiated synapses 25 A Synaptic state after STDP 20 15 10 5 0 B 50 100 150 200 Spiking order 250 Figure 5: Compensating for variability. A Some synapses (dots) become potentiated (light) while others remain depressed (dark) after STDP. B The number of potentiated synapses neurons make (pluses) and receive (circles) is negatively (r = -0.71) and positively (r = 0.76) correlated to their rank in the spiking order, respectively. Comparing the number of potentiated synapses each neuron made or received with its excitability conﬁrmed the PEP hypothesis (i.e., leading neurons provide additional synaptic current to lagging neurons via potentiated recurrent synapses). In this experiment, to eliminate variability due to noise (as opposed to excitability), we provided a 17 by 17 cluster of neurons with a regular 200Hz excitatory input. Theta inhibition was present as before and all synapses were initialized to the depressed state. After 10 seconds of STDP, a large fraction of the synapses were potentiated (Figure 5A). When the number of potentiated synapses each neuron made or received was plotted versus its rank in spiking order (Figure 5B), a clear correlation emerged (r = -0.71 or 0.76, respectively). As expected, neurons that spiked early made more and received fewer potentiated synapses. In contrast, neurons that spiked late made fewer and received more potentiated synapses. 5 Pattern Completion After STDP, we found that the network could recall an entire pattern given a subset, thus the same mechanisms that compensated for variability and noise could also compensate for lack of information. We chose a 9 by 9 cluster of neurons as our pattern and delivered a poisson-like spike train with mean rate of 67Hz to each one as in the ﬁrst experiment. Theta inhibition was present as before and all synapses were initialized to the depressed state. Before STDP, we stimulated a subset of the pattern and only neurons in that subset spiked (Figure 6A). After ﬁve seconds of STDP, we stimulated the same subset again. This time they recruited spikes from other neurons in the pattern, completing it (Figure 6B). Upon varying the fraction of the pattern presented, we found that the fraction recalled increased faster than the fraction presented. We selected subsets of the original pattern randomly, varying the fraction of neurons chosen from 0.1 to 1.0 (ten trials for each). We classiﬁed neurons as active if they spiked in the two second period over which we recorded. Thus, we characterized PEP’s pattern-recall performance as a function of the probability that the pattern in question’s neurons are activated (Figure 6C). At a fraction of 0.50 presented, nearly all of the neurons in the pattern are consistently activated (0.91±0.06), showing robust pattern completion. We ﬁtted the recall performance with a sigmoid that reached 0.50 recall fraction with an input fraction of 0.30. No spurious neurons were activated during any trials. Rate(Hz) Rate(Hz) 8 7 7 6 6 5 5 0.6 0.4 2 0.2 0 0 3 3 2 1 1 A 0.8 4 4 Network activity before STDP 1 Fraction of pattern actived 8 0 B Network activity after STDP C 0 0.2 0.4 0.6 0.8 Fraction of pattern stimulated 1 Figure 6: Associative recall. A Before STDP, half of the neurons in a pattern are stimulated; only they are activated. B After STDP, half of the neurons in a pattern are stimulated, and all are activated. C The fraction of the pattern activated grows faster than the fraction stimulated. 6 Discussion Our results demonstrate that PEP successfully compensates for graded variations in our silicon recurrent network using binary (on–off) synapses (in contrast with [8], where weights are graded). While our chip results are encouraging, variability was not eliminated in every case. In the case of the lowest input (50Hz), we see virtually no change (Figure 4A). We suspect the timing remains imprecise because, with such low input, neurons do not spike every theta cycle and, consequently, provide fewer opportunities for the STDP synapses to potentiate. This shortfall illustrates the system’s limits; it can only compensate for variability within certain bounds, and only for activity appropriate to the PEP model. As expected, STDP is the mechanism responsible for PEP. STDP potentiated recurrent synapses from leading neurons to lagging neurons, reducing the disparity among the diverse population of neurons. Even though the STDP circuits are themselves variable, with different efﬁcacies and time constants, when using timing the sign of the weight-change is always correct (data not shown). For this reason, we chose STDP over other more physiological implementations of plasticity, such as membrane-voltage-dependent plasticity (MVDP), which has the capability to learn with graded voltage signals [9], such as those found in active dendrites, providing more computational power [10]. Previously, we investigated a MVDP circuit, which modeled a voltage-dependent NMDAreceptor-gated synapse [11]. It potentiated when the calcium current analog exceeded a threshold, which was designed to occur only during a dendritic action potential. This circuit produced behavior similar to STDP, implying it could be used in PEP. However, it was sensitive to variability in the NMDA and potentiation thresholds, causing a fraction of the population to potentiate anytime the synapse received an input and another fraction to never potentiate, rendering both subpopulations useless. Therefore, the simpler, less biophysical STDP circuit won out over the MVDP circuit: In our system timing is everything. Associative storage and recall naturally emerge in the PEP network when synapses between neurons coactivated by a pattern are potentiated. These synapses allow neurons to recruit their peers when a subset of the pattern is presented, thereby completing the pattern. However, this form of pattern storage and completion differs from Hopﬁeld’s attractor model [12] . Rather than forming symmetric, recurrent neuronal circuits, our recurrent network forms asymmetric circuits in which neurons make connections exclusively to less excitable neurons in the pattern. In both the poisson-like and regular cases (Figures 4 & 5), only about six percent of potentiated connections were reciprocated, as expected by chance. We plan to investigate the storage capacity of this asymmetric form of associative memory. Our system lends itself to modeling brain regions that use precise spike timing, such as the hippocampus. We plan to extend the work presented to store and recall sequences of patterns, as the hippocampus is hypothesized to do. Place cells that represent different locations spike at different phases of the theta cycle, in relation to the distance to their preferred locations. This sequential spiking will allow us to link patterns representing different locations in the order those locations are visited, thereby realizing episodic memory. We propose PEP as a candidate neural mechanism for information coding and storage in the hippocampal system. Observations from the CA1 region of the hippocampus suggest that basal dendrites (which primarily receive excitation from recurrent connections) support submillisecond timing precision, consistent with PEP [13]. We have shown, in a silicon model, PEP’s ability to exploit such fast recurrent connections to sharpen timing precision as well as to associatively store and recall patterns. Acknowledgments We thank Joe Lin for assistance with chip generation. The Ofﬁce of Naval Research funded this work (Award No. N000140210468). References [1] O’Keefe J. & Recce M.L. (1993). Phase relationship between hippocampal place units and the EEG theta rhythm. Hippocampus 3(3):317-330. [2] Mehta M.R., Lee A.K. & Wilson M.A. (2002) Role of experience and oscillations in transforming a rate code into a temporal code. Nature 417(6890):741-746. [3] Bi G.Q. & Wang H.X. (2002) Temporal asymmetry in spike timing-dependent synaptic plasticity. Physiology & Behavior 77:551-555. [4] Rodriguez-Vazquez, A., Linan, G., Espejo S. & Dominguez-Castro R. (2003) Mismatch-induced trade-offs and scalability of analog preprocessing visual microprocessor chips. Analog Integrated Circuits and Signal Processing 37:73-83. [5] Boahen K.A. (2000) Point-to-point connectivity between neuromorphic chips using address events. IEEE Transactions on Circuits and Systems II 47:416-434. [6] Culurciello E.R., Etienne-Cummings R. & Boahen K.A. (2003) A biomorphic digital image sensor. IEEE Journal of Solid State Circuits 38:281-294. [7] Boﬁll A., Murray A.F & Thompson D.P. (2005) Citcuits for VLSI Implementation of Temporally Asymmetric Hebbian Learning. In: Advances in Neural Information Processing Systems 14, MIT Press, 2002. [8] Cameron K., Boonsobhak V., Murray A. & Renshaw D. (2005) Spike timing dependent plasticity (STDP) can ameliorate process variations in neuromorphic VLSI. IEEE Transactions on Neural Networks 16(6):1626-1627. [9] Chicca E., Badoni D., Dante V., D’Andreagiovanni M., Salina G., Carota L., Fusi S. & Del Giudice P. (2003) A VLSI recurrent network of integrate-and-ﬁre neurons connected by plastic synapses with long-term memory. IEEE Transaction on Neural Networks 14(5):1297-1307. [10] Poirazi P., & Mel B.W. (2001) Impact of active dendrites and structural plasticity on the memory capacity of neural tissue. Neuron 29(3)779-796. [11] Arthur J.V. & Boahen K. (2004) Recurrently connected silicon neurons with active dendrites for one-shot learning. In: IEEE International Joint Conference on Neural Networks 3, pp.1699-1704. [12] Hopﬁeld J.J. (1984) Neurons with graded response have collective computational properties like those of two-state neurons. Proceedings of the National Academy of Science 81(10):3088-3092. [13] Ariav G., Polsky A. & Schiller J. (2003) Submillisecond precision of the input-output transformation function mediated by fast sodium dendritic spikes in basal dendrites of CA1 pyramidal neurons. Journal of Neuroscience 23(21):7750-7758.

Reference: text

Summary: the most important sentenses genereted by tfidf model

sentIndex sentText sentNum sentScore

1 edu Abstract We describe a neuromorphic chip that uses binary synapses with spike timing-dependent plasticity (STDP) to learn stimulated patterns of activity and to compensate for variability in excitability. [sent-4, score-0.955]

2 Speciﬁcally, STDP preferentially potentiates (turns on) synapses that project from excitable neurons, which spike early, to lethargic neurons, which spike late. [sent-5, score-0.891]

3 The additional excitatory synaptic current makes lethargic neurons spike earlier, thereby causing neurons that belong to the same pattern to spike in synchrony. [sent-6, score-1.279]

4 1 Variability in Neural Systems Evidence suggests precise spike timing is important in neural coding, speciﬁcally, in the hippocampus. [sent-8, score-0.495]

5 The hippocampus uses timing in the spike activity of place cells (in addition to rate) to encode location in space [1]. [sent-9, score-0.625]

6 Place cells employ a phase code: the timing at which a neuron spikes relative to the phase of the inhibitory theta rhythm (5-12Hz) conveys information. [sent-10, score-0.756]

7 As an animal approaches a place cell’s preferred location, the place cell not only increases its spike rate, but also spikes at earlier phases in the theta cycle. [sent-11, score-0.746]

8 To implement a phase code, the theta rhythm is thought to prevent spiking until the input synaptic current exceeds the sum of the neuron threshold and the decreasing inhibition on the downward phase of the cycle [2]. [sent-12, score-0.693]

9 However, even with identical inputs and common theta inhibition, neurons do not spike in synchrony. [sent-13, score-0.825]

10 Lethargic neurons (such as those with high thresholds) spike late in the theta cycle, since their input exceeds the sum of the neuron threshold and theta inhibition only after the theta inhibition has had time to decrease. [sent-15, score-1.643]

11 Conversely, excitable neurons (such as those with low thresholds) spike early in the theta cycle. [sent-16, score-0.863]

12 Consequently, variability in excitability translates into variability in timing. [sent-17, score-0.306]

13 We hypothesize that the hippocampus achieves its precise spike timing (about 10ms) through plasticity enhanced phase-coding (PEP). [sent-18, score-0.648]

14 The source of hippocampal timing precision in the presence of variability (and noise) remains unexplained. [sent-19, score-0.38]

15 Synaptic plasticity can compensate for variability in excitability if it increases excitatory synaptic input to neurons in inverse proportion to their excitabilities. [sent-20, score-0.655]

16 Recasting this in a phase-coding framework, we desire a learning rule that increases excitatory synaptic input to neurons directly related to their phases. [sent-21, score-0.355]

17 Neurons that lag require additional synaptic input, whereas neurons that lead 120µm 190µm A B Figure 1: STDP Chip. [sent-22, score-0.284]

18 The spike timing-dependent plasticity (STDP) observed in the hippocampus satisﬁes this requirement [3]. [sent-26, score-0.471]

19 It requires repeated pre-before-post spike pairings (within a time window) to potentiate and repeated post-before-pre pairings to depress a synapse. [sent-27, score-0.556]

20 It has a 32 by 32 array of excitatory principal neurons commingled with a 16 by 16 array of inhibitory interneurons that are not used here (Figure 1A). [sent-37, score-0.348]

21 The address-event representation (AER) [5] is used to transmit spikes off chip and to receive afferent and recurrent spike input. [sent-39, score-0.633]

22 To conﬁgure the STDP Chip as a recurrent network, we embedded it in a circuit board (Figure 1B). [sent-40, score-0.315]

23 The DACs control the analog biases in the system, including the leak current, which the PC varies in real-time to create the global inhibitory theta rhythm. [sent-45, score-0.423]

24 The principal neuron consists of a refractory period and calcium-dependent potassium circuit (RCK), a synapse circuit, and a soma circuit (Figure 2A). [sent-46, score-0.765]

25 B Spikes (dots) from 81 principal neurons are temporally dispersed, when excited by poisson-like inputs (58Hz) and inhibited by the common 8. [sent-61, score-0.263]

26 The soma is a modiﬁed version of the LPF, which receives additional input from an axonhillock circuit (AH). [sent-65, score-0.287]

27 The synapse circuit is similar to RCK in structure but differs in function: It is activated not by the principal neuron itself but by the STDP circuits (or directly by afferent spikes that bypass these circuits, i. [sent-76, score-0.597]

28 It receives excitation from the synapse circuit and shunting inhibition from RCK and has a leak current as well. [sent-81, score-0.409]

29 STDP circuit ~LTP SRAM Presynaptic spike A ~LTD Inverse number of pairings Integrator Decay Postsynaptic spike Potentiation 0. [sent-85, score-0.877]

30 1 Depression -80 -40 0 Presynaptic spike Postsynaptic spike 40 Spike timing: t pre - t post (ms) 80 B Figure 3: STDP circuit design and characterization. [sent-89, score-0.807]

31 B The circuit potentiates when the presynaptic spike precedes the postsynaptic spike and depresses when the postsynaptic spike precedes the presynaptic spike. [sent-91, score-1.4]

32 The soma circuit is connected to an AH, the locus of spike generation. [sent-92, score-0.576]

33 It initiates the AER signaling process required to send a spike off chip. [sent-94, score-0.318]

34 To characterize principal neuron variability, we excited 81 neurons with poisson-like 58Hz spike trains (Figure 2B). [sent-95, score-0.652]

35 We made these spike trains poisson-like by starting with a regular 200Hz spike train and dropping spikes randomly, with probability of 0. [sent-96, score-0.717]

36 Thus spikes were delivered to neurons that won the coin toss in synchrony every 5ms. [sent-98, score-0.358]

37 However, neurons did not lock onto the input synchrony due to ﬁltering by the synaptic time constant (see Figure 2B). [sent-99, score-0.345]

38 They also received a common inhibitory input at the theta frequency (8. [sent-100, score-0.405]

39 Each neuron was prevented from ﬁring more than one spike in a theta cycle by its model calcium-dependent potassium channel population. [sent-102, score-0.786]

40 To quantify the spike variability, we used timing precision, which we deﬁne as twice the standard deviation of spike times accumulated from ﬁve theta cycles. [sent-104, score-1.106]

41 With an input rate of 58Hz the timing precision was 34ms. [sent-105, score-0.251]

42 Its capacitor is charged when that spike occurs and discharges linearly thereafter. [sent-111, score-0.367]

43 A postsynaptic spike samples the charge remaining on the capacitor, passes it through an exponential function, and dumps the resultant charge into the integrator. [sent-112, score-0.512]

44 If it exceeds a threshold, the SRAM switches state from depressed to potentiated (∼LTD goes high and ∼LTP goes low). [sent-115, score-0.309]

45 The depression side of the STDP circuit is exactly symmetric, except that it responds to postsynaptic activation followed by presynaptic activation and switches the SRAM’s state from potentiated to depressed (∼LTP goes high and ∼LTD goes low). [sent-116, score-0.649]

46 When the SRAM is in the potentiated state, the presynaptic 50 After STDP 83 92 100 Timing precision(ms) Before STDP 75 B Before STDP After STDP 40 30 20 10 0 50 60 70 80 90 Input rate(Hz) 100 50 58 67 text A 0. [sent-117, score-0.238]

47 A Spike rasters of 81 neurons (9 by 9 cluster) display synchrony over a two-fold range of input rates after STDP. [sent-124, score-0.275]

48 C Each neuron (center box) sends synapses to (dark gray) and receives synapses from (light gray) twenty-one randomly chosen neighbors up to ﬁve nodes away (black indicates both connections). [sent-126, score-0.359]

49 spike activates the principal neuron’s synapse; otherwise the spike has no effect. [sent-127, score-0.685]

50 We characterized the STDP circuit by activating a plastic synapse and a ﬁxed synapse– which elicits a spike at different relative times. [sent-128, score-0.603]

51 4 Recurrent Network We carried out an experiment designed to test the STDP circuit’s ability to compensate for variability in spike timing through PEP. [sent-137, score-0.664]

52 Each neuron received recurrent connections from 21 randomly selected neurons within an 11 by 11 neighborhood centered on itself (see Figure 4C). [sent-138, score-0.479]

53 Conversely, it made recurrent connections to randomly chosen neurons within the same neighborhood. [sent-139, score-0.368]

54 We chose a 9 by 9 cluster of neurons and delivered spikes at a mean rate of 50 to 100Hz to each one (dropping spikes with a probability of 0. [sent-141, score-0.407]

55 5 from a regular 200Hz train) and provided common theta inhibition as before. [sent-143, score-0.359]

56 We compared the variability in spike timing after ﬁve seconds of learning with the initial distribution. [sent-144, score-0.621]

57 Before STDP, spike timing among neurons was highly variable (except for the very highest input rate). [sent-146, score-0.738]

58 A Some synapses (dots) become potentiated (light) while others remain depressed (dark) after STDP. [sent-151, score-0.393]

59 B The number of potentiated synapses neurons make (pluses) and receive (circles) is negatively (r = -0. [sent-152, score-0.531]

60 Comparing the number of potentiated synapses each neuron made or received with its excitability conﬁrmed the PEP hypothesis (i. [sent-155, score-0.482]

61 , leading neurons provide additional synaptic current to lagging neurons via potentiated recurrent synapses). [sent-157, score-0.78]

62 In this experiment, to eliminate variability due to noise (as opposed to excitability), we provided a 17 by 17 cluster of neurons with a regular 200Hz excitatory input. [sent-158, score-0.382]

63 Theta inhibition was present as before and all synapses were initialized to the depressed state. [sent-159, score-0.286]

64 After 10 seconds of STDP, a large fraction of the synapses were potentiated (Figure 5A). [sent-160, score-0.385]

65 When the number of potentiated synapses each neuron made or received was plotted versus its rank in spiking order (Figure 5B), a clear correlation emerged (r = -0. [sent-161, score-0.461]

66 As expected, neurons that spiked early made more and received fewer potentiated synapses. [sent-164, score-0.473]

67 In contrast, neurons that spiked late made fewer and received more potentiated synapses. [sent-165, score-0.473]

68 5 Pattern Completion After STDP, we found that the network could recall an entire pattern given a subset, thus the same mechanisms that compensated for variability and noise could also compensate for lack of information. [sent-166, score-0.255]

69 We chose a 9 by 9 cluster of neurons as our pattern and delivered a poisson-like spike train with mean rate of 67Hz to each one as in the ﬁrst experiment. [sent-167, score-0.622]

70 Theta inhibition was present as before and all synapses were initialized to the depressed state. [sent-168, score-0.286]

71 Before STDP, we stimulated a subset of the pattern and only neurons in that subset spiked (Figure 6A). [sent-169, score-0.377]

72 This time they recruited spikes from other neurons in the pattern, completing it (Figure 6B). [sent-171, score-0.295]

73 Upon varying the fraction of the pattern presented, we found that the fraction recalled increased faster than the fraction presented. [sent-172, score-0.292]

74 We selected subsets of the original pattern randomly, varying the fraction of neurons chosen from 0. [sent-173, score-0.341]

75 We classiﬁed neurons as active if they spiked in the two second period over which we recorded. [sent-176, score-0.26]

76 Thus, we characterized PEP’s pattern-recall performance as a function of the probability that the pattern in question’s neurons are activated (Figure 6C). [sent-177, score-0.316]

77 50 presented, nearly all of the neurons in the pattern are consistently activated (0. [sent-179, score-0.316]

78 A Before STDP, half of the neurons in a pattern are stimulated; only they are activated. [sent-195, score-0.273]

79 B After STDP, half of the neurons in a pattern are stimulated, and all are activated. [sent-196, score-0.273]

80 C The fraction of the pattern activated grows faster than the fraction stimulated. [sent-197, score-0.238]

81 6 Discussion Our results demonstrate that PEP successfully compensates for graded variations in our silicon recurrent network using binary (on–off) synapses (in contrast with [8], where weights are graded). [sent-198, score-0.363]

82 While our chip results are encouraging, variability was not eliminated in every case. [sent-199, score-0.251]

83 We suspect the timing remains imprecise because, with such low input, neurons do not spike every theta cycle and, consequently, provide fewer opportunities for the STDP synapses to potentiate. [sent-201, score-1.186]

84 STDP potentiated recurrent synapses from leading neurons to lagging neurons, reducing the disparity among the diverse population of neurons. [sent-204, score-0.64]

85 Even though the STDP circuits are themselves variable, with different efﬁcacies and time constants, when using timing the sign of the weight-change is always correct (data not shown). [sent-205, score-0.245]

86 It potentiated when the calcium current analog exceeded a threshold, which was designed to occur only during a dendritic action potential. [sent-208, score-0.233]

87 However, it was sensitive to variability in the NMDA and potentiation thresholds, causing a fraction of the population to potentiate anytime the synapse received an input and another fraction to never potentiate, rendering both subpopulations useless. [sent-210, score-0.579]

88 Therefore, the simpler, less biophysical STDP circuit won out over the MVDP circuit: In our system timing is everything. [sent-211, score-0.348]

89 Associative storage and recall naturally emerge in the PEP network when synapses between neurons coactivated by a pattern are potentiated. [sent-212, score-0.475]

90 These synapses allow neurons to recruit their peers when a subset of the pattern is presented, thereby completing the pattern. [sent-213, score-0.417]

91 Rather than forming symmetric, recurrent neuronal circuits, our recurrent network forms asymmetric circuits in which neurons make connections exclusively to less excitable neurons in the pattern. [sent-215, score-0.797]

92 In both the poisson-like and regular cases (Figures 4 & 5), only about six percent of potentiated connections were reciprocated, as expected by chance. [sent-216, score-0.218]

93 Our system lends itself to modeling brain regions that use precise spike timing, such as the hippocampus. [sent-218, score-0.318]

94 Place cells that represent different locations spike at different phases of the theta cycle, in relation to the distance to their preferred locations. [sent-220, score-0.611]

95 Observations from the CA1 region of the hippocampus suggest that basal dendrites (which primarily receive excitation from recurrent connections) support submillisecond timing precision, consistent with PEP [13]. [sent-223, score-0.437]

96 We have shown, in a silicon model, PEP’s ability to exploit such fast recurrent connections to sharpen timing precision as well as to associatively store and recall patterns. [sent-224, score-0.467]

97 Phase relationship between hippocampal place units and the EEG theta rhythm. [sent-232, score-0.352]

98 (2005) Spike timing dependent plasticity (STDP) can ameliorate process variations in neuromorphic VLSI. [sent-275, score-0.291]

99 (2003) A VLSI recurrent network of integrate-and-ﬁre neurons connected by plastic synapses with long-term memory. [sent-285, score-0.467]

100 (2004) Recurrently connected silicon neurons with active dendrites for one-shot learning. [sent-295, score-0.324]

similar papers computed by tfidf model

tfidf for this paper:

wordName wordTfidf (topN-words)

[('stdp', 0.53), ('spike', 0.318), ('theta', 0.293), ('neurons', 0.214), ('timing', 0.177), ('potentiated', 0.173), ('circuit', 0.171), ('pep', 0.153), ('synapses', 0.144), ('variability', 0.126), ('chip', 0.125), ('lpf', 0.115), ('synapse', 0.114), ('recurrent', 0.109), ('rck', 0.102), ('sram', 0.102), ('pe', 0.097), ('ah', 0.088), ('soma', 0.087), ('spikes', 0.081), ('plasticity', 0.077), ('depressed', 0.076), ('hippocampus', 0.076), ('potentiate', 0.073), ('neuron', 0.071), ('synaptic', 0.07), ('pairings', 0.07), ('circuits', 0.068), ('fraction', 0.068), ('charge', 0.068), ('inhibition', 0.066), ('presynaptic', 0.065), ('silicon', 0.064), ('potassium', 0.064), ('potentiation', 0.061), ('pattern', 0.059), ('ishunt', 0.059), ('stimulated', 0.058), ('postsynaptic', 0.058), ('leak', 0.058), ('integrator', 0.054), ('excitability', 0.054), ('boahen', 0.051), ('principal', 0.049), ('capacitor', 0.049), ('spiked', 0.046), ('graded', 0.046), ('dendrites', 0.046), ('depression', 0.046), ('precision', 0.045), ('connections', 0.045), ('dacs', 0.044), ('isoma', 0.044), ('lethargic', 0.044), ('mvdp', 0.044), ('inhibitory', 0.043), ('activated', 0.043), ('compensate', 0.043), ('excitatory', 0.042), ('received', 0.04), ('cycle', 0.04), ('decay', 0.039), ('cpld', 0.038), ('refractory', 0.038), ('pmos', 0.038), ('excitable', 0.038), ('neuromorphic', 0.037), ('board', 0.035), ('usb', 0.035), ('ltd', 0.035), ('cacy', 0.034), ('spiking', 0.033), ('synchrony', 0.032), ('hippocampal', 0.032), ('associative', 0.032), ('storage', 0.031), ('aer', 0.031), ('delivered', 0.031), ('transistor', 0.031), ('calcium', 0.031), ('ltp', 0.031), ('rhythm', 0.031), ('goes', 0.03), ('phase', 0.03), ('voltage', 0.03), ('input', 0.029), ('extender', 0.029), ('potentiates', 0.029), ('recalled', 0.029), ('submillisecond', 0.029), ('vdd', 0.029), ('ve', 0.029), ('ram', 0.029), ('analog', 0.029), ('place', 0.027), ('recall', 0.027), ('activity', 0.027), ('isyn', 0.025), ('subcircuits', 0.025), ('depress', 0.025)]

similar papers list:

simIndex simValue paperId paperTitle

same-paper 1 0.99999917 118 nips-2005-Learning in Silicon: Timing is Everything

Author: John V. Arthur, Kwabena Boahen

2 0.45266697 8 nips-2005-A Criterion for the Convergence of Learning with Spike Timing Dependent Plasticity

Author: Robert A. Legenstein, Wolfgang Maass

Abstract: We investigate under what conditions a neuron can learn by experimentally supported rules for spike timing dependent plasticity (STDP) to predict the arrival times of strong “teacher inputs” to the same neuron. It turns out that in contrast to the famous Perceptron Convergence Theorem, which predicts convergence of the perceptron learning rule for a simpliﬁed neuron model whenever a stable solution exists, no equally strong convergence guarantee can be given for spiking neurons with STDP. But we derive a criterion on the statistical dependency structure of input spike trains which characterizes exactly when learning with STDP will converge on average for a simple model of a spiking neuron. This criterion is reminiscent of the linear separability criterion of the Perceptron Convergence Theorem, but it applies here to the rows of a correlation matrix related to the spike inputs. In addition we show through computer simulations for more realistic neuron models that the resulting analytically predicted positive learning results not only hold for the common interpretation of STDP where STDP changes the weights of synapses, but also for a more realistic interpretation suggested by experimental data where STDP modulates the initial release probability of dynamic synapses. 1

3 0.33569574 181 nips-2005-Spiking Inputs to a Winner-take-all Network

Author: Matthias Oster, Shih-Chii Liu

Abstract: Recurrent networks that perform a winner-take-all computation have been studied extensively. Although some of these studies include spiking networks, they consider only analog input rates. We present results of this winner-take-all computation on a network of integrate-and-ﬁre neurons which receives spike trains as inputs. We show how we can conﬁgure the connectivity in the network so that the winner is selected after a pre-determined number of input spikes. We discuss spiking inputs with both regular frequencies and Poisson-distributed rates. The robustness of the computation was tested by implementing the winner-take-all network on an analog VLSI array of 64 integrate-and-ﬁre neurons which have an innate variance in their operating parameters. 1

4 0.2977024 39 nips-2005-Beyond Pair-Based STDP: a Phenomenological Rule for Spike Triplet and Frequency Effects

Author: Jean-pascal Pfister, Wulfram Gerstner

Abstract: While classical experiments on spike-timing dependent plasticity analyzed synaptic changes as a function of the timing of pairs of pre- and postsynaptic spikes, more recent experiments also point to the effect of spike triplets. Here we develop a mathematical framework that allows us to characterize timing based learning rules. Moreover, we identify a candidate learning rule with ﬁve variables (and 5 free parameters) that captures a variety of experimental data, including the dependence of potentiation and depression upon pre- and postsynaptic ﬁring frequencies. The relation to the Bienenstock-Cooper-Munro rule as well as to some timing-based rules is discussed. 1

5 0.29464906 188 nips-2005-Temporally changing synaptic plasticity

Author: Minija Tamosiunaite, Bernd Porr, Florentin Wörgötter

Abstract: Recent experimental results suggest that dendritic and back-propagating spikes can inﬂuence synaptic plasticity in different ways [1]. In this study we investigate how these signals could temporally interact at dendrites leading to changing plasticity properties at local synapse clusters. Similar to a previous study [2], we employ a differential Hebbian plasticity rule to emulate spike-timing dependent plasticity. We use dendritic (D-) and back-propagating (BP-) spikes as post-synaptic signals in the learning rule and investigate how their interaction will inﬂuence plasticity. We will analyze a situation where synapse plasticity characteristics change in the course of time, depending on the type of post-synaptic activity momentarily elicited. Starting with weak synapses, which only elicit local D-spikes, a slow, unspeciﬁc growth process is induced. As soon as the soma begins to spike this process is replaced by fast synaptic changes as the consequence of the much stronger and sharper BP-spike, which now dominates the plasticity rule. This way a winner-take-all-mechanism emerges in a two-stage process, enhancing the best-correlated inputs. These results suggest that synaptic plasticity is a temporal changing process by which the computational properties of dendrites or complete neurons can be substantially augmented. 1

6 0.22244443 99 nips-2005-Integrate-and-Fire models with adaptation are good enough

7 0.2039305 157 nips-2005-Principles of real-time computing with feedback applied to cortical microcircuit models

8 0.14986299 1 nips-2005-AER Building Blocks for Multi-Layer Multi-Chip Neuromorphic Vision Systems

9 0.14893976 64 nips-2005-Efficient estimation of hidden state dynamics from spike trains

10 0.12406213 164 nips-2005-Representing Part-Whole Relationships in Recurrent Neural Networks

11 0.1167304 22 nips-2005-An Analog Visual Pre-Processing Processor Employing Cyclic Line Access in Only-Nearest-Neighbor-Interconnects Architecture

12 0.110456 67 nips-2005-Extracting Dynamical Structure Embedded in Neural Activity

13 0.10698571 124 nips-2005-Measuring Shared Information and Coordinated Activity in Neuronal Networks

14 0.10185814 176 nips-2005-Silicon growth cones map silicon retina

15 0.10140388 61 nips-2005-Dynamical Synapses Give Rise to a Power-Law Distribution of Neuronal Avalanches

16 0.096354611 106 nips-2005-Large-scale biophysical parameter estimation in single neurons via constrained linear regression

17 0.091259569 40 nips-2005-CMOL CrossNets: Possible Neuromorphic Nanoelectronic Circuits

18 0.07879474 134 nips-2005-Neural mechanisms of contrast dependent receptive field size in V1

19 0.068448529 25 nips-2005-An aVLSI Cricket Ear Model

20 0.067337304 129 nips-2005-Modeling Neural Population Spiking Activity with Gibbs Distributions

similar papers computed by lsi model

lsi for this paper:

topicId topicWeight

[(0, 0.172), (1, -0.534), (2, -0.16), (3, -0.225), (4, -0.038), (5, -0.04), (6, 0.042), (7, 0.124), (8, 0.044), (9, -0.081), (10, -0.046), (11, 0.003), (12, 0.021), (13, 0.116), (14, -0.012), (15, -0.083), (16, -0.085), (17, -0.016), (18, 0.011), (19, -0.075), (20, -0.01), (21, -0.022), (22, -0.027), (23, -0.016), (24, 0.014), (25, 0.001), (26, -0.013), (27, 0.022), (28, -0.009), (29, 0.034), (30, -0.063), (31, 0.035), (32, 0.019), (33, -0.066), (34, -0.032), (35, 0.042), (36, -0.015), (37, -0.018), (38, 0.016), (39, -0.037), (40, 0.019), (41, -0.064), (42, -0.05), (43, 0.058), (44, -0.028), (45, 0.019), (46, 0.043), (47, -0.004), (48, -0.009), (49, 0.042)]

similar papers list:

simIndex simValue paperId paperTitle

same-paper 1 0.98129475 118 nips-2005-Learning in Silicon: Timing is Everything

Author: John V. Arthur, Kwabena Boahen

2 0.84727007 8 nips-2005-A Criterion for the Convergence of Learning with Spike Timing Dependent Plasticity

Author: Robert A. Legenstein, Wolfgang Maass

3 0.81867182 39 nips-2005-Beyond Pair-Based STDP: a Phenomenological Rule for Spike Triplet and Frequency Effects

Author: Jean-pascal Pfister, Wulfram Gerstner

4 0.81607395 181 nips-2005-Spiking Inputs to a Winner-take-all Network

Author: Matthias Oster, Shih-Chii Liu

5 0.77581018 188 nips-2005-Temporally changing synaptic plasticity

Author: Minija Tamosiunaite, Bernd Porr, Florentin Wörgötter

6 0.73757017 99 nips-2005-Integrate-and-Fire models with adaptation are good enough

7 0.52307475 176 nips-2005-Silicon growth cones map silicon retina

8 0.51498377 64 nips-2005-Efficient estimation of hidden state dynamics from spike trains

9 0.50855899 157 nips-2005-Principles of real-time computing with feedback applied to cortical microcircuit models

10 0.48356637 61 nips-2005-Dynamical Synapses Give Rise to a Power-Law Distribution of Neuronal Avalanches

11 0.48268285 1 nips-2005-AER Building Blocks for Multi-Layer Multi-Chip Neuromorphic Vision Systems

12 0.46762395 40 nips-2005-CMOL CrossNets: Possible Neuromorphic Nanoelectronic Circuits

13 0.33430353 165 nips-2005-Response Analysis of Neuronal Population with Synaptic Depression

14 0.32828221 128 nips-2005-Modeling Memory Transfer and Saving in Cerebellar Motor Learning

15 0.32765338 106 nips-2005-Large-scale biophysical parameter estimation in single neurons via constrained linear regression

16 0.32617068 22 nips-2005-An Analog Visual Pre-Processing Processor Employing Cyclic Line Access in Only-Nearest-Neighbor-Interconnects Architecture

17 0.31596988 164 nips-2005-Representing Part-Whole Relationships in Recurrent Neural Networks

18 0.30960953 67 nips-2005-Extracting Dynamical Structure Embedded in Neural Activity

19 0.30133134 124 nips-2005-Measuring Shared Information and Coordinated Activity in Neuronal Networks

20 0.2803725 25 nips-2005-An aVLSI Cricket Ear Model

similar papers computed by lda model

lda for this paper:

topicId topicWeight

[(3, 0.016), (10, 0.024), (27, 0.022), (31, 0.038), (34, 0.028), (39, 0.022), (44, 0.011), (55, 0.018), (57, 0.07), (60, 0.021), (65, 0.011), (69, 0.069), (71, 0.01), (73, 0.018), (88, 0.041), (91, 0.456), (99, 0.021)]

similar papers list:

simIndex simValue paperId paperTitle

same-paper 1 0.95708841 118 nips-2005-Learning in Silicon: Timing is Everything

Author: John V. Arthur, Kwabena Boahen

2 0.94411772 38 nips-2005-Beyond Gaussian Processes: On the Distributions of Infinite Networks

Author: Ricky Der, Daniel D. Lee

Abstract: A general analysis of the limiting distribution of neural network functions is performed, with emphasis on non-Gaussian limits. We show that with i.i.d. symmetric stable output weights, and more generally with weights distributed from the normal domain of attraction of a stable variable, that the neural functions converge in distribution to stable processes. Conditions are also investigated under which Gaussian limits do occur when the weights are independent but not identically distributed. Some particularly tractable classes of stable distributions are examined, and the possibility of learning with such processes.

3 0.92652458 202 nips-2005-Variational EM Algorithms for Non-Gaussian Latent Variable Models

Author: Jason Palmer, Kenneth Kreutz-Delgado, Bhaskar D. Rao, David P. Wipf

Abstract: We consider criteria for variational representations of non-Gaussian latent variables, and derive variational EM algorithms in general form. We establish a general equivalence among convex bounding methods, evidence based methods, and ensemble learning/Variational Bayes methods, which has previously been demonstrated only for particular cases.

4 0.83784145 76 nips-2005-From Batch to Transductive Online Learning

Author: Sham Kakade, Adam Tauman Kalai

Abstract: It is well-known that everything that is learnable in the difﬁcult online setting, where an arbitrary sequences of examples must be labeled one at a time, is also learnable in the batch setting, where examples are drawn independently from a distribution. We show a result in the opposite direction. We give an efﬁcient conversion algorithm from batch to online that is transductive: it uses future unlabeled data. This demonstrates the equivalence between what is properly and efﬁciently learnable in a batch model and a transductive online model.

5 0.6251272 201 nips-2005-Variational Bayesian Stochastic Complexity of Mixture Models

Author: Kazuho Watanabe, Sumio Watanabe

Abstract: The Variational Bayesian framework has been widely used to approximate the Bayesian learning. In various applications, it has provided computational tractability and good generalization performance. In this paper, we discuss the Variational Bayesian learning of the mixture of exponential families and provide some additional theoretical support by deriving the asymptotic form of the stochastic complexity. The stochastic complexity, which corresponds to the minimum free energy and a lower bound of the marginal likelihood, is a key quantity for model selection. It also enables us to discuss the eﬀect of hyperparameters and the accuracy of the Variational Bayesian approach as an approximation of the true Bayesian learning. 1

6 0.60107088 181 nips-2005-Spiking Inputs to a Winner-take-all Network

7 0.59659106 197 nips-2005-Unbiased Estimator of Shape Parameter for Spiking Irregularities under Changing Environments

8 0.57647735 96 nips-2005-Inference with Minimal Communication: a Decision-Theoretic Variational Approach

9 0.55842936 90 nips-2005-Hot Coupling: A Particle Approach to Inference and Normalization on Pairwise Undirected Graphs

10 0.55759227 157 nips-2005-Principles of real-time computing with feedback applied to cortical microcircuit models

11 0.53701717 32 nips-2005-Augmented Rescorla-Wagner and Maximum Likelihood Estimation

12 0.53160322 205 nips-2005-Worst-Case Bounds for Gaussian Process Models

13 0.5249958 61 nips-2005-Dynamical Synapses Give Rise to a Power-Law Distribution of Neuronal Avalanches

14 0.51751006 30 nips-2005-Assessing Approximations for Gaussian Process Classification

15 0.51545393 52 nips-2005-Correlated Topic Models

16 0.5069443 54 nips-2005-Data-Driven Online to Batch Conversions

17 0.4905256 43 nips-2005-Comparing the Effects of Different Weight Distributions on Finding Sparse Representations

18 0.48319578 124 nips-2005-Measuring Shared Information and Coordinated Activity in Neuronal Networks

19 0.47628605 168 nips-2005-Rodeo: Sparse Nonparametric Regression in High Dimensions

20 0.47305977 67 nips-2005-Extracting Dynamical Structure Embedded in Neural Activity