Drop, Swap, and Generate: A Self-Supervised Approach for Generating Neural Activity
This work addresses the challenge of interpreting neural circuits for neuroscientists, though it is incremental as it builds on existing generative modeling techniques.
The authors tackled the problem of learning meaningful representations of neural activity without labels by introducing Swap-VAE, an unsupervised method that uses dropout and time jittering to create transformed views, resulting in disentangled representations linked to behavior in primate brain data.
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.