Timezone: »

 
Oral
Drop, Swap, and Generate: A Self-Supervised Approach for Generating Neural Activity
Ran Liu · Mehdi Azabou · Max Dabagia · Chi-Heng Lin · Mohammad Gheshlaghi Azar · Keith Hengen · Michal Valko · Eva Dyer

Fri Dec 10 04:20 PM -- 04:35 PM (PST) @
in Oral Session 5: Generative Modeling »

Meaningful and simplified representations of neural activity can yield insights into how and what information is being processed within a neural circuit. However, without labels, finding representations that reveal the link between the brain and behavior can be challenging. Here, we introduce a novel unsupervised approach for learning disentangled representations of neural activity called Swap-VAE. Our approach combines a generative modeling framework with an instance-specific alignment loss that tries to maximize the representational similarity between transformed views of the input (brain state). These transformed (or augmented) views are created by dropping out neurons and jittering samples in time, which intuitively should lead the network to a representation that maintains both temporal consistency and invariance to the specific neurons used to represent the neural state. Through evaluations on both synthetic data and neural recordings from hundreds of neurons in different primate brains, we show that it is possible to build representations that disentangle neural datasets along relevant latent dimensions linked to behavior.

Keywords: Self-Supervised Learning · Generative Model · Representation Learning

Author Information

Ran Liu (Georgia Institute of Technology)

I am a 4th year Ph.D. student in the Machine Learning Program at Georgia Tech. I conduct my research in the Neural Data Science Lab advised by Prof. Eva Dyer. My research interests lie at the intersection of Machine (Deep) Learning, Computational Neuroscience, and Computer Vision.

Mehdi Azabou (Georgia Institute of Technology)
Max Dabagia (Georgia Institute of Technology)
Chi-Heng Lin (gatech)
Mohammad Gheshlaghi Azar (DeepMind)
Keith Hengen (Washington University, St. Louis)
Michal Valko (DeepMind Paris / Inria / ENS Paris-Saclay)
Michal Valko

Michal is a machine learning scientist in DeepMind Paris, tenured researcher at Inria, and the lecturer of the master course Graphs in Machine Learning at l'ENS Paris-Saclay. Michal is primarily interested in designing algorithms that would require as little human supervision as possible. This means 1) reducing the “intelligence” that humans need to input into the system and 2) minimizing the data that humans need to spend inspecting, classifying, or “tuning” the algorithms. That is why he is working on methods and settings that are able to deal with minimal feedback, such as deep reinforcement learning, bandit algorithms, or self-supervised learning. Michal is actively working on represenation learning and building worlds models. He is also working on deep (reinforcement) learning algorithm that have some theoretical underpinning. He has also worked on sequential algorithms with structured decisions where exploiting the structure leads to provably faster learning. He received his Ph.D. in 2011 from the University of Pittsburgh under the supervision of Miloš Hauskrecht and after was a postdoc of Rémi Munos before taking a permanent position at Inria in 2012.

Eva Dyer (Georgia Institute of Technology)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors