Compositional generalization through abstract representations in human and artificial neural networks
Nov 28, 2022
Speakers
Takuya Ito
Speaker · 0 followers
Tim Klinger
Speaker · 0 followers
Douglas H. Schultz
Speaker · 0 followers
John D. Murray
Speaker · 0 followers
Michael W. Cole
Speaker · 0 followers
Mattia Rigotti
Speaker · 0 followers
About
Humans have a remarkable ability to rapidly generalize to new tasks that is difficult to reproduce in artificial learning systems.Compositionality has been proposed as a key mechanism supporting generalization in humans, but evidence of its neural implementation and impact on behavior is still scarce. Here we study the computational properties associated with compositional generalization in both humans and artificial neural networks (ANNs) on a highly compositional task. First, we identified behavioral signatures of compositional generalization in humans, along with their neural correlates using whole-cortex functional magnetic resonance imaging (fMRI) data. Next, we designed pretraining paradigms aided by a procedure we term primitives pretraining to endow compositional task elements into ANNs. We found that ANNs with this prior knowledge had greater correspondence with human behavior and neural compositional signatures. Importantly, primitives pretraining induced abstract internal representations, excellent zero-shot generalization, and sample-efficient learning. Moreover, it gave rise to a hierarchy of abstract representations that matched human fMRI data, where sensory rule abstractions emerged in early sensory areas, and motor rule abstractions emerged in later motor areas. Our findings give empirical support to the role of compositional generalization in humans behavior, implicate abstract representations as its neural implementation, and illustrate that these representations can be embedded into ANNs by designing simple and efficient pretraining procedures.Humans have a remarkable ability to rapidly generalize to new tasks that is difficult to reproduce in artificial learning systems.Compositionality has been proposed as a key mechanism supporting generalization in humans, but evidence of its neural implementation and impact on behavior is still scarce. Here we study the computational properties associated with compositional generalization in both humans and artificial neural networks (ANNs) on a highly compositional task. First, we identified beh…
Organizer
NeurIPS 2022
Account · 952 followers
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
A Geometric Perspective on Variational Autoencoders
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Unlabelled Sample Compression Schemes for Intersection-Closed Classes and Extremal Classes
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Stochastic Oracles and How to Define Them
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Multi-objective Tree-structured Parzen Estimator Meets Meta-learning
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Fair Multimodal Checklists for Interpretable Clinical Time Series Prediction
Watch later
Qixuan Jin, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Outlier Suppression: Pushing the Limit of Low-bit Transformer Language Models
Watch later
Xiuying Wei, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%