Information bottleneck method

Résumé

The information bottleneck method is a technique in information theory introduced by Naftali Tishby, Fernando C. Pereira, and William Bialek. It is designed for finding the best tradeoff between accuracy and complexity (compression) when summarizing (e.g. clustering) a random variable X, given a joint probability distribution p(X,Y) between X and an observed relevant variable Y - and self-described as providing "a surprisingly rich framework for discussing a variety of problems in signal processing and learning". Applications include distributional clustering and dimension reduction, and more recently it has been suggested as a theoretical foundation for deep learning. It generalized the classical notion of minimal sufficient statistics from parametric statistics to arbitrary distributions, not necessarily of exponential form. It does so by relaxing the sufficiency condition to capture some fraction of the mutual information with the relevant variable Y. The informati

Source officielle

https://en.wikipedia.org/wiki/Information_bottleneck_method

À propos de ce résultat

Cette page est générée automatiquement et peut contenir des informations qui ne sont pas correctes, complètes, à jour ou pertinentes par rapport à votre recherche. Il en va de même pour toutes les autres pages de ce site. Veillez à vérifier les informations auprès des sources officielles de l'EPFL.

Publications associées (18)

Chargement

Afficher plus

Publications associées

Chargement

Personnes associées (4)

Personnes associées

Chargement

Unités associées (1)

Unités associées

Chargement

Concepts associés

Chargement

Cours associés (1)

The Deep Learning for NLP course provides an overview of neural network based methods applied to text. The focus is on models particularly suited to the properties of human language, such as categorical, unbounded, and structured representations, and very large input and output vocabularies.

Cours associés

Chargement

Séances de cours associées (2)

Séances de cours associées

Chargement

Publications associées (18)

An Information Theoretic Approach to Speaker Diarization of Meeting Recordings

Deepu Vijayasenan

In this thesis we investigate a non parametric approach to speaker diarization for meeting recordings based on an information theoretic framework. The problem is formulated using the Information Bottleneck (IB) principle. Unlike other approaches where the distance between speaker segments is arbitrarily introduced, the IB method seeks the partition that maximizes the mutual information between observations and variables relevant for the problem while minimizing the distortion between observations. The distance between speech segments is selected as the Jensen-Shannon divergence as it arises from the IB objective function optimization. In the first part of the thesis, we explore IB based diarization with Mel frequency cepstral coefficients (MFCC) as input features. We study issues related to IB based speaker diarization such as optimizing the IB objective function, criteria for inferring the number of speakers. Furthermore, we benchmark the proposed system against a state-of-the-art systemon the NIST RT06 (Rich Transcription) meeting data for speaker diarization. The IB based system achieves similar speaker error rates (16.8%) as compared to a baseline HMM/GMM system (17.0%). This approach being non parametric clustering, perform diarization six times faster than realtime while the baseline is slower than realtime. The second part of thesis proposes a novel feature combination system in the context of IB diarization. Both speaker clustering and speaker realignment steps are discussed. In contrary to current systems, the proposed method avoids the feature combination by averaging log-likelihood scores. Two different sets of features were considered – (a) combination of MFCC features with time delay of arrival features (b) a four feature stream combination that combines MFCC, TDOA, modulation spectrum and frequency domain linear prediction. Experiments show that the proposed system achieve 5% absolute improvement over the baseline in case of two feature combination, and 7% in case of four feature combination. The increase in algorithm complexity of the IB system is minimal with more features. The system with four feature input performs in real time that is ten times faster than the GMM based system.

EPFL2010

Chargement

Function Identification in Neuron Populations via Information Bottleneck

Sreenivas Kartik Buddha, Michael Christoph Gastpar

It is plausible to hypothesize that the spiking responses of certain neurons represent functions of the spiking signals of other neurons. A natural ensuing question concerns how to use experimental data to infer what kind of a function is being computed. Model-based approaches typically require assumptions on how information is represented. By contrast, information measures are sensitive only to relative behavior: information is unchanged by applying arbitrary invertible transformations to the involved random variables. This paper develops an approach based on the information bottleneck method that attempts to find such functional relationships in a neuron population. Specifically, the information bottleneck method is used to provide appropriate compact representations which can then be parsed to infer functional relationships. In the present paper, the parsing step is specialized to the case of remapped-linear functions. The approach is validated on artificial data and then applied to recordings from the motor cortex of a macaque monkey performing an arm-reaching task. Functional relationships are identified and shown to exhibit some degree of persistence across multiple trials of the same experiment.

MDPI AG2013

Chargement

, ,

Large-scale models for learning fixed-dimensional cross-lingual sentence representations like LASER (Artetxe and Schwenk, 2019b) lead to significant improvement in performance on downstream tasks. However, further increases and modifications based on such large-scale models are usually impractical due to memory limitations. In this work, we introduce a lightweight dual-transformer architecture with just 2 layers for generating memory-efficient cross-lingual sentence representations. We explore different training tasks and observe that current cross-lingual training tasks leave a lot to be desired for this shallow architecture. To ameliorate this, we propose a novel cross-lingual language model, which combines the existing single-word masked language model with the newly proposed cross-lingual token-level reconstruction task. We further augment the training task by the introduction of two computationally-lite sentence-level contrastive learning tasks to enhance the alignment of cross-lingual sentence representation space, which compensates for the learning bottleneck of the lightweight transformer for generative tasks. Our comparisons with competing models on cross-lingual sentence retrieval and multilingual document classification confirm the effectiveness of the newly proposed training tasks for a shallow model.

ASSOC COMPUTATIONAL LINGUISTICS-ACL2021

Chargement

Afficher plus

An Information Theoretic Approach to Speaker Diarization of Meeting Recordings

Deepu Vijayasenan

EPFL2010