Timezone: »
A key goal of unsupervised learning is to go beyond density estimation and sample generation to reveal the structure inherent within observed data. Such structure can be expressed in the pattern of interactions between explanatory latent variables captured through a probabilistic graphical model. Although the learning of structured graphical models has a long history, much recent work in unsupervised modelling has instead emphasised flexible deep-network-based generation, either transforming independent latent generators to model complex data or assuming that distinct observed variables are derived from different latent nodes. Here, we extend amortised variational inference to incorporate structured factors over multiple variables, able to capture the observation-induced posterior dependence between latents that results from “explaining away” and thus allow complex observations to depend on multiple nodes of a structured graph. We show that appropriately parametrised factors can be combined efficiently with variational message passing in rich graphical structures. We instantiate the framework in nonlinear Gaussian Process Factor Analysis, evaluating the structured recognition framework using synthetic data from known generative processes. We fit the GPFA model to high-dimensional neural spike data from the hippocampus of freely moving rodents, where the model successfully identifies latent signals that correlate with behavioural covariates.
Author Information
Changmin Yu (UCL)
Hugo Soulat (Gatsby Unit)
Neil Burgess (University College London)
Maneesh Sahani (Gatsby Unit, UCL)
More from the Same Authors
-
2021 Spotlight: Probabilistic Tensor Decomposition of Neural Population Spiking Activity »
Hugo Soulat · Sepiedeh Keshavarzi · Troy Margrie · Maneesh Sahani -
2022 : Leveraging Episodic Memory to Improve World Models for Reinforcement Learning »
Julian Coda-Forno · Changmin Yu · Qinghai Guo · Zafeirios Fountas · Neil Burgess -
2022 : Constructing Memory: Consolidation as Teacher-Student Training of a Generative Model »
Eleanor Spens · Neil Burgess -
2023 Poster: Successor-Predecessor Intrinsic Exploration »
Changmin Yu · Neil Burgess · Samuel J Gershman · Maneesh Sahani -
2023 Poster: A State Representation for Diminishing Rewards »
Ted Moskovitz · Samo Hromadka · Ahmed Touati · Diana Borsa · Maneesh Sahani -
2021 Poster: Probabilistic Tensor Decomposition of Neural Population Spiking Activity »
Hugo Soulat · Sepiedeh Keshavarzi · Troy Margrie · Maneesh Sahani -
2020 Poster: Non-reversible Gaussian processes for identifying latent dynamical structure in neural data »
Virginia Rutten · Alberto Bernacchia · Maneesh Sahani · Guillaume Hennequin -
2020 Oral: Non-reversible Gaussian processes for identifying latent dynamical structure in neural data »
Virginia Rutten · Alberto Bernacchia · Maneesh Sahani · Guillaume Hennequin -
2020 Poster: Organizing recurrent network dynamics by task-computation to enable continual learning »
Lea Duncker · Laura N Driscoll · Krishna V Shenoy · Maneesh Sahani · David Sussillo -
2019 Poster: A neurally plausible model for online recognition and postdiction in a dynamical environment »
Li Kevin Wenliang · Maneesh Sahani -
2019 Poster: A neurally plausible model learns successor representations in partially observable environments »
Eszter Vértes · Maneesh Sahani -
2019 Poster: Coordinated hippocampal-entorhinal replay as structural inference »
Talfan Evans · Neil Burgess -
2019 Oral: A neurally plausible model learns successor representations in partially observable environments »
Eszter Vértes · Maneesh Sahani -
2019 Poster: Kernel Instrumental Variable Regression »
Rahul Singh · Maneesh Sahani · Arthur Gretton -
2019 Oral: Kernel Instrumental Variable Regression »
Rahul Singh · Maneesh Sahani · Arthur Gretton -
2018 Poster: Flexible and accurate inference and learning for deep generative models »
Eszter Vértes · Maneesh Sahani -
2018 Poster: Temporal alignment and latent Gaussian process factor inference in population spike trains »
Lea Duncker · Maneesh Sahani -
2015 Poster: Bayesian Manifold Learning: The Locally Linear Latent Variable Model (LL-LVM) »
Mijung Park · Wittawat Jitkrittum · Ahmad Qamar · Zoltan Szabo · Lars Buesing · Maneesh Sahani -
2013 Workshop: Acquiring and Analyzing the Activity of Large Neural Ensembles »
Srinivas C Turaga · Lars Buesing · Maneesh Sahani · Jakob H Macke -
2013 Poster: Extracting regions of interest from biological images with convolutional sparse block coding »
Marius Pachitariu · Adam M Packer · Noah Pettit · Henry Dalgleish · Michael Hausser · Maneesh Sahani -
2013 Poster: Recurrent linear models of simultaneously-recorded neural populations »
Marius Pachitariu · Biljana Petreska · Maneesh Sahani -
2013 Spotlight: Recurrent linear models of simultaneously-recorded neural populations »
Marius Pachitariu · Biljana Petreska · Maneesh Sahani -
2012 Poster: Spectral learning of linear dynamics from generalised-linear observations with application to neural population data »
Lars Buesing · Jakob H Macke · Maneesh Sahani -
2012 Oral: Spectral learning of linear dynamics from generalised-linear observations with application to neural population data »
Lars Buesing · Jakob H Macke · Maneesh Sahani -
2012 Poster: Learning visual motion in recurrent neural networks »
Marius Pachitariu · Maneesh Sahani -
2011 Oral: Empirical models of spiking in neural populations »
Jakob H Macke · Lars Buesing · John P Cunningham · Byron M Yu · Krishna V Shenoy · Maneesh Sahani -
2011 Poster: Empirical models of spiking in neural populations »
Jakob H Macke · Lars Buesing · John P Cunningham · Byron M Yu · Krishna V Shenoy · Maneesh Sahani -
2011 Poster: Dynamical segmentation of single trials from population neural data »
Biljana Petreska · Byron M Yu · John P Cunningham · Gopal Santhanam · Stephen I Ryu · Krishna V Shenoy · Maneesh Sahani -
2011 Poster: Probabilistic amplitude and frequency demodulation »
Richard Turner · Maneesh Sahani -
2011 Spotlight: Probabilistic amplitude and frequency demodulation »
Richard Turner · Maneesh Sahani -
2010 Session: The Sam Roweis Symposium »
Maneesh Sahani -
2009 Poster: Occlusive Components Analysis »
Jörg Lücke · Richard Turner · Maneesh Sahani · Marc Henniges -
2008 Poster: Gaussian-process factor analysis for low-dimensional single-trial analysis of neural population activity »
Byron M Yu · John P Cunningham · Gopal Santhanam · Stephen I Ryu · Krishna V Shenoy · Maneesh Sahani -
2007 Workshop: Beyond Simple Cells: Probabilistic Models for Visual Cortical Processing »
Richard Turner · Pietro Berkes · Maneesh Sahani -
2007 Oral: Inferring Elapsed Time from Stochastic Neural Processes »
Misha B Ahrens · Maneesh Sahani -
2007 Spotlight: Inferring Neural Firing Rates from Spike Trains Using Gaussian Processes »
John P Cunningham · Byron M Yu · Krishna V Shenoy · Maneesh Sahani -
2007 Poster: Inferring Neural Firing Rates from Spike Trains Using Gaussian Processes »
John P Cunningham · Byron M Yu · Krishna V Shenoy · Maneesh Sahani -
2007 Poster: Inferring Elapsed Time from Stochastic Neural Processes »
Misha B Ahrens · Maneesh Sahani -
2007 Poster: Modeling Natural Sounds with Modulation Cascade Processes »
Richard Turner · Maneesh Sahani -
2007 Poster: On Sparsity and Overcompleteness in Image Models »
Pietro Berkes · Richard Turner · Maneesh Sahani