The functional specialization of visual cortex emerges from training parallel pathways with self-supervised learning
Dec 6, 2021
Speakers
Shahab Bakhtiari
Speaker · 0 followers
Patrick Mineault
Speaker · 0 followers
Timothy Lillicrap
Speaker · 8 followers
Christopher Pack
Speaker · 0 followers
Blake Richards
Speaker · 1 follower
About
The visual system of mammals is comprised of parallel, hierarchical specialized pathways. Different pathways are specialized in so far as they use representations that are more suitable for supporting specific downstream behaviours. In particular, the clearest example is the specialization of the ventral ("what") and dorsal ("where") pathways of the visual cortex. These two pathways support behaviours related to visual recognition and movement, respectively. To-date, deep neural networks have mostly been used as models of the ventral, recognition pathway. However, it is unknown whether both pathways can be modelled with a single deep ANN. Here, we ask whether a single model with a single loss function can capture the properties of both the ventral and the dorsal pathways. We explore this question using data from mice, who like other mammals, have specialized pathways that appear to support recognition and movement behaviours. We show that when we train a deep neural network architecture with two parallel pathways using a self-supervised predictive loss function, we can outperform other models in fitting mouse visual cortex. Moreover, we can model both the dorsal and ventral pathways. These results demonstrate that a self-supervised predictive learning approach applied to parallel pathway architectures can account for some of the functional specialization seen in mammalian visual systems.The visual system of mammals is comprised of parallel, hierarchical specialized pathways. Different pathways are specialized in so far as they use representations that are more suitable for supporting specific downstream behaviours. In particular, the clearest example is the specialization of the ventral ("what") and dorsal ("where") pathways of the visual cortex. These two pathways support behaviours related to visual recognition and movement, respectively. To-date, deep neural networks have mo…
Organizer
NeurIPS 2021
Account · 1.9k followers
About NeurIPS 2021
Neural Information Processing Systems (NeurIPS) is a multi-track machine learning and computational neuroscience conference that includes invited talks, demonstrations, symposia and oral and poster presentations of refereed papers. Following the conference, there are workshops which provide a less formal setting.
Like the format? Trust SlidesLive to capture your next event!
Professional recording and live streaming, delivered globally.
Sharing
Recommended Videos
Presentations on similar topic, category or speaker
Robust Robotic Control from Pixels using Contrastive Recurrent State-Space Models
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Multiple Instance Learning for Brain Tumor Detection with Magnetic Resonance Spectroscopy Data
Watch later
Diyuan Lu, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Numerical Composition of Differential Privacy
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Implicit Regularization in Matrix Sensing via Mirror Descent
Watch later
Fan Wu, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Affordance Learning from Play for Sample-Efficient Policy Learning
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Fair Exploration via Axiomatic Bargaining
Watch later
Jackie Baek, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%