Large-Scale Content-Based Audio Retrieval from Text Queries (2009)
Gal Chechik, Samy Bengio, Eugene Ie, Dick Lyon, Martin Rehn
In content-based audio retrieval, the goal is to find sound recordings (audio documents) based on their acoustic features. This content-based approach differs from retrieval approaches that index...
LETTER Communicated by David Horn Spike-Timing-Dependent Plasticity and Relevant Mutual (2009)
Information Maximization, Gal Chechik
Synaptic plasticity was recently shown to depend on the relative timing of the pre- and postsynaptic spikes. This article analytically derives a spike-dependent learning rule based on the principle...
Discrete profile comparison using information bottleneck (2009)
Gal Chechik, Robin Friedman, Eleazar Eskin
Sequence homologs are an important source of information about proteins. Amino acid profiles, representing the position-specific mutation probabilities found in profiles, are a richer encoding of...
Filling missing components in yeast metabolic pathways (2009)
Gal Chechik, Aviv Regev, Daphne Koller
using heterogeneous data
Max-margin classification of data with absent features (2009)
Gal Chechik, Geremy Heitz, Gal Elidan, Pieter Abbeel
We consider the problem of learning classifiers in structured domains, where some objects have a subset of features that are inherently absent due to complex relationships between the features....
1 Temporally Dependent Plasticity: An Information Theoretic Account (2008)
Abstract The paradigm of Hebbian learning has recently received a novel interpretation with the discovery of synaptic plasticity that depends on the relative timing of pre and post synaptic spikes....
1 Temporally Dependent Plasticity: An Information Theoretic Account (2008)
The paradigm of Hebbian learning has recently received a novel interpretation with the discovery of synaptic plasticity that depends on the relative timing of pre and post synaptic spikes. This paper...
Embedding Heterogeneous Data using Statistical Models (2008)
Amir Globerson, Gal Chechik, Fernando Pereira, Naftali Tishby
Embedding algorithms are a method for revealing low dimensional structure in complex data. Most embedding algorithms are designed to handle objects of a single type for which pairwise distances are...
Gal Chechik, Michael J. Anderson, Omer Bar-yosef, Eric D. Young, Naftali Tishby, Israel Nelken
Information processing by a sensory system is reflected in the changes in stimulus representation along its successive processing stages. We measured information content and stimulus-induced...
Separation of overlapping subpopulations by mutual information (2008)
Identifying ancestral sequences is an important first step in understanding population history and dynamics. However, several interesting cases including human genetic variation feature highly...
NIPS workshop on New Problems and Methods in Computational Biology (2007)
Chechik, Gal, Leslie, Christina, Noble, William, Rätsch, Gunnar, Morris, Quaid, Tsuda, Koji
Since its conception half a century ago, Hebbian learning has become a fundamental paradigm in the neurosciences. The idea that neurons that re together wire together has become fairly well...
Correspondence should be addressed to: (2007)
Isaac Meilijson, Gal Chechik, Gal Chechik
In this paper we revisit the classical neuroscience paradigm of Hebbian learning. We nd that it is dicult to achieve effective associative memory storage by Hebbian synaptic learning, since it...
1 Temporally Dependent Plasticity: An Information Theoretic Account (2007)
The paradigm of Hebbian learning has recently received a novel interpretation with the discovery of synaptic plasticity that depends on the relative timing of pre and post synaptic spikes. This paper...
Neuronal Regulation Implements Efficient Synaptic Pruning (2007)
Gal Chechik, Isaac Meilijson, Eytan Ruppin
Human and animal studies show that mammalian brain undergoes massive synaptic pruning during childhood, removing about half of the synapses until puberty. We have previously shown that maintaining...
Gal Chechik, Isaac Meilijson, Eytan Ruppin
This paper studies the fundamental interplay between Hebbian synaptic changes and neuronally driven processes modifying synaptic efficacies, and its role in associative memory learning. The...
Correspondence should be addressed to: (2007)
Isaac Meilijson, Gal Chechik, Gal Chechik
In this paper we revisit the classical neuroscience paradigm of Hebbian learning. We nd that it is dicult to achieve effective associative memory storage by Hebbian synaptic learning, since it...
Eective Learning Requires Neuronal Remodeling of Hebbian Synapses (2007)
This paper revisits the classical neuroscience paradigm of Hebbian learning. We nd that a necessary requirement for eective associative memory learning is that the ecacies of the incoming synapses...
Schools of Medicine and Mathematical Sciences (2007)
Neuronal regulation is a mechanism that was recently found to maintain the homeostasis of the neuron's membrane potential. We show that the operation of this mechanism may lead to a bi-modal...
"Doctor of Philosophy" (2007)
Someone once said that every manuscript is just a graphomaniac appendix to the acknowledgment page. The current page provides this hypothesis with hard empirical evidence. This thesis summarizes a...
Euclidean Embedding of Co-occurrence Data (2007)
Globerson, Amir, Chechik, Gal, Pereira, Fernando, Tishby, Professor Naftali
Embedding algorithms search for a low dimensional continuous representation of data, but most algorithms only handle objects of a single type for which pairwise distances are specified. This paper...
Max-margin classification of incomplete data (2007)
We consider the problem of learning classifiers for structurally incomplete data, where some objects have a subset of features inherently absent due to complex relationships between the features. The...
Max-margin classification of incomplete data (2007)
We consider the problem of learning classifiers for structurally incomplete data, where some objects have a subset of features inherently absent due to complex relationships between the features. The...
Max-margin classification of incomplete data (2007)
Gal Chechik, Geremy Heitz, Gal Elidan, Pieter Abbeel, Daphne Koller
We consider the problem of learning classifiers for structurally incomplete data, where some objects have a subset of features inherently absent due to complex relationships between the features. The...
Discrete profile comparison using information bottleneck (2006)
O'Rourke, Sean, Chechik, Gal, Friedman, Robin, Eskin, Eleazar
Abstract Sequence homologs are an important source of information about proteins. Amino acid profiles, representing the position-specific mutation probabilities found in profiles, are a richer...
Temporal and cross-subject probabilistic models for fmri prediction tasks (2006)
Alexis Battle, Gal Chechik, Daphne Koller
We present a probabilistic model applied to the fMRI video rating prediction task of the Pittsburgh Brain Activity Interpretation Competition (PBAIC) [2]. Our goal is to predict a time series of...
Temporal and cross-subject probabilistic models for fmri prediction tasks (2006)
Alexis Battle, Gal Chechik, Daphne Koller
We present a probabilistic model applied to the fMRI video rating prediction task of the Pittsburgh Brain Activity Interpretation Competition (PBAIC) [2]. Our goal is to predict a time series of...
Euclidean embedding of co-occurrence data (2005)
Amir Globerson, Gal Chechik, Fernando Pereira, Naftali Tishby
Abstract Embedding algorithms search for low dimensional structure in complexdata, but most algorithms only handle objects of a single type for which pairwise distances are specified. This paper...
Euclidean embedding of co-occurrence data (2005)
Amir Globerson, Gal Chechik, Fernando Pereira, Naftali Tishby
Embedding algorithms search for low dimensional structure in complex data, but most algorithms only handle objects of a single type for which pairwise distances are specified. This paper describes a...
Euclidean embedding of co-occurrence data (2005)
Amir Globerson, Gal Chechik, Fernando Pereira, Naftali Tishby
Embedding algorithms search for low dimensional structure in complex data, but most algorithms only handle objects of a single type for which pairwise distances are specified. This paper describes a...
Euclidean embedding of co-occurrence data (2005)
Amir Globerson, Gal Chechik, Fernando Pereira, Naftali Tishby, John Lafferty
Embedding algorithms search for a low dimensional continuous representation of data, but most algorithms only handle objects of a single type for which pairwise distances are specified. This paper...
Israel Nelken, Gal Chechik, Andrew J. King
Abstract. Neurons can transmit information about sensory stimuli via their firing rate, spike latency, or by the occurrence of complex spike patterns. Identifying which aspects of the neural...
A needle in a haystack: Local one-class optimization (2004)
This paper addresses the problem of finding a small and coherent subset of points in a given data. This problem, sometimes referred to as one-class or set covering, requires to find a small-radius...
A needle in a haystack: Local one-class optimization (2004)
This paper addresses the problem of finding a small and coherent subset of points in a given data. This problem, sometimes referred to as one-class or set covering, requires to find a small-radius...
A needle in a haystack: Local one-class optimization (2004)
This paper addresses the problem of finding a small and coherent subset of points in a given data. This problem, sometimes referred to as one-class or set covering, requires to find a small-radius...
A needle in a haystack: Local one-class optimization (2004)
This paper addresses the problem of finding a small and coherent subset of points in a given data. This problem, sometimes referred to as one-class or set covering, requires to find a small-radius...
Spike-timing dependent plasticity and relevant mutual information maximization (2003)
Synaptic plasticity was recently shown to depend on the relative timing of the pre and post synaptic spikes. The current paper analytically derives a spike dependent learning rule based on the...
Information bottleneck for gaussian variables (2003)
Gal Chechik, Amir Globerson, Naftali Tishby, Yair Weiss
∗ Both authors contributed equally The problem of extracting the relevant aspects of data was addressed through the information bottleneck (IB) method, by (soft) clustering one variable while...
This work was carried out under the supervision of Prof. Naftali Tishby and Dr. Israel
Are there representations in embodied evolved agents? taking measures (2003)
Abstract. The question of conceptual representation has received considerable attention in philosophy, neuroscience and embodied evolved agents. Numerous theories on the interpretation of the term...
Sufficient dimensionality reduction with irrelevant statistics (2003)
Amir Globerson, Gal Chechik, Naftali Tishby
The problem of unsupervised dimensionality reduction of stochastic variables while preserving their most relevant characteristics is fundamental for the analysis of complex data. Unfortunately, this...
Are there representations in embodied evolved agents? taking measures (2003)
Abstract. The question of conceptual representation has received considerable attention in philosophy, neuroscience and embodied evolved agents. Numerous theories on the interpretation of the term...
Spike-timing dependent plasticity and relevant mutual information maximization (2003)
Synaptic plasticity was recently shown to depend on the relative timing of the pre and post synaptic spikes. The current paper analytically derives a spike dependent learning rule based on the...
Information Bottleneck and Linear Projections of Gaussian Processes (2003)
The problem of extracting the relevant aspects of data was addressed for categorical variables through the information bottleneck (IB) method, by compressing one variable while preserving information...
Information bottleneck for gaussian variables (2003)
Gal Chechik, Amir Globerson, Naftali Tishby, Yair Weiss
∗ Both authors contributed equally The problem of extracting the relevant aspects of data was addressed through the information bottleneck (IB) method, by (soft) clustering one variable while...
This work was carried out under the supervision of Prof. Naftali Tishby and Dr. Israel
Information Bottleneck for Gaussian Variables (2003)
Gal Chechik, Amir Globerson, Naftali Tishby, Yair Weiss
The problem of extracting the relevant aspects of data was addressed through the information bottleneck (IB) method, by (soft) clustering one variable while preserving information about another -...
Information Bottleneck for Gaussian Variables (2003)
Gal Chechik, Amir Globerson, Naftali Tishby, Yair Weiss
The problem of extracting the relevant aspects of data was addressed through the information bottleneck (IB) method, by (soft) clustering one variable while preserving information about another -...
Sufficient Dimensionality Reduction with Irrelevance Statistics (2003)
Amir Globerson, Gal Chechik, Naftali Tishby
The problem of unsupervised dimensionality reduction of stochastic variables while preserving their most relevant characteristics is fundamental for the analysis of complex data. Unfortunately, this...
Extracting relevant structures with side information (2002)
The problem of extracting the relevant aspects of data, in face of multiple conflicting structures, is inherent to modeling of complex data. Extracting structure in one random variable that is...
Extracting relevant structures with side information (2002)
The problem of extracting the relevant aspects of data, in face of multiple conflicting structures, is inherent to modeling of complex data. Extracting structure in one random variable that is...
Group redundancy measures reveal redundancy reduction in the auditory pathway (2002)
Gal Chechik, Amir Globerson, Naftali Tishby, Michael J. Anderson, Eric D. Young, Israel Nelken
The way groups of auditory neurons interact to code acoustic information is investigated using an information theoretic approach. We develop measures of redundancy among groups of neurons, and apply...
Group redundancy measures reveal redundancy reduction in the auditory pathway (2002)
Gal Chechik, Amir Globerson, Naftali Tishby, Michael J. Anderson, Eric D. Young, Israel Nelken
The way groups of auditory neurons interact to code acoustic information is investigated using an information theoretic approach. Identifying the case of stimulus-conditioned independent neurons, we...
Types, Super-Types and the Mutual (2002)
Discovering stochastic dependencies in empirical data is a fundamental problem in data modeling. Dependencies are usually measured using the mutual information or chi square statistics, and then have...
Extracting relevant structures with side information (2002)
The problem of extracting the relevant aspects of data, in face of multiple conflicting structures, is inherent to modeling of complex data. Extracting structure in one random variable that is...
Distributional clustering of movements based on neural responses. unpublished manuscript (2001)
Amir Globerson, Gal Chechik, Naftali Tishby, Orna Steinberg, Eilon Vaadia
The generation and control of movement by the brain is a highly complicated phenomenon, and the neural mechanisms underlying it are still largely unknown. This is certainly true for complex movements...
Temporal Dependent Plasticity: An Information Theoretic Account (2001)
The fundamental paradigm of Hebbian learning has recently received a novel interpretation with the discovery of synaptic plasticity that depends on the relative timing of pre and post synaptic...
Temporally dependent plasticity: An information theoretic account (2000)
The paradigm of Hebbian learning has recently received a novel interpretation with the discovery of synaptic plasticity that depends on the relative timing of pre and post synaptic spikes. This paper...
Correspondence should be addressed to: (2000)
Gal Chechik, Isaac Meilijson, Gal Chechik
0 In this paper we revisit the classical neuroscience paradigm of Hebbian learning. We find that it is difficult to achieve ef-fective associative memory storage by Hebbian synaptic learn-ing, since...
Effective learning requires neuronal remodeling of hebbian synapses (1999)
This paper revisits the classical neuroscience paradigm of Hebbian learning. We find that a necessary requirement for effective associative memory learning is that the efficacies of the incoming...
Neuronal regulation: A mechanism for synaptic pruning during brain maturation (1999)
Human and animal studies show that mammalian brains undergoes massive synaptic pruning during childhood, removing about half of the synapses until puberty. We have previously shown that maintaining...
Effective And Optimal Storage of Memory Patterns With Variable Coding Levels (1999)
This paper studies the storage of memory patterns with varying coding levels (fraction of firing neurons within a pattern) in an associative memory network. It was previously shown that effective...
Beyond Hebbian plasticity: Effective learning with ineffective Hebbian learning rules (1999)
In this paper we revisit the classical neuroscience paradigm of Hebbian learning. We find that a necessary requirement for effective associative memory learning is that the efficacies of the incoming...
Synaptic pruning in development: A novel account in neural terms (1998)
Research in humans and primates shows that the developmental course of the brain involves synaptic over-growth followed by marked selective pruning. Previous explanations have suggested that this...
Synaptic pruning in development: a computational account (1998)
Research in humans and primates shows that the developmental course of the brain involves synaptic over-growth followed by marked selective pruning. Previous explanations have suggested that this...
Synaptic Pruning in Development: A Novel Account in Neural Terms (1998)
Gal Chechik, Isaac Meilijson, Eytan Ruppin
This paper shows that in the general modeling framework of associative memory networks this explanation does not hold. We put forward a different explanation: the observed profile of synaptic density...
Neuronal Regulation: A Mechanism For Synaptic Pruning During Brain Maturation (1998)
Gal Chechik, Isaac Meilijson, Eytan Ruppin
Human and animal studies show that mammalian brains undergoes massive synaptic pruning during childhood, removing about half of the synapses until puberty. We have previously shown that maintaining...
Computational Aspects of Synaptic Elimination (1997)
The Raymond, Beverly Sackler, Gal Chechik, Prof Isaco Meilijson, Jonathan Cohen, Eyal Cohen (zoro
Research in humans and primates shows that the developmental course of the brain involves synaptic over-growth followed by marked selective pruning which eliminates about half of the synapses of the...
Discrete profile comparison using information bottleneck
O'Rourke, Sean, Chechik, Gal, Friedman, Robin, Eskin, Eleazar
Sequence homologs are an important source of information about proteins. Amino acid profiles, representing the position-specific mutation probabilities found in profiles, are a richer encoding of...