Timezone: »

 
Poster
Using natural language and program abstractions to instill human inductive biases in machines
Sreejan Kumar · Carlos G. Correa · Ishita Dasgupta · Raja Marjieh · Michael Y Hu · Robert Hawkins · Jonathan D Cohen · nathaniel daw · Karthik Narasimhan · Tom Griffiths

Thu Dec 01 02:00 PM -- 04:00 PM (PST) @ Hall J #942
in Poster Session 6 »

Strong inductive biases give humans the ability to quickly learn to perform a variety of tasks. Although meta-learning is a method to endow neural networks with useful inductive biases, agents trained by meta-learning may sometimes acquire very different strategies from humans. We show that co-training these agents on predicting representations from natural language task descriptions and programs induced to generate such tasks guides them toward more human-like inductive biases. Human-generated language descriptions and program induction models that add new learned primitives both contain abstract concepts that can compress description length. Co-training on these representations result in more human-like behavior in downstream meta-reinforcement learning agents than less abstract controls (synthetic language descriptions, program induction without learned primitives), suggesting that the abstraction supported by these representations is key.

Keywords: Cognitive Science · Reinforcement Learning · Program Induction · Meta-Learning · natural language · human intelligence
[ Paper]  [ Poster]  [ OpenReview] 

Author Information

Sreejan Kumar (Princeton University)
Carlos G. Correa (Princeton University)
Carlos G. Correa

I'm a grad student at the Princeton Neuroscience Institute, studying decision making and planning by using computational models to predict human behavior. I'm advised by Nathaniel Daw and Tom Griffiths.

Ishita Dasgupta (DeepMind)
Raja Marjieh (Princeton University)
Michael Y Hu (Princeton University)
Robert Hawkins (Princeton University)
Jonathan D Cohen (Princeton University)
nathaniel daw (Princeton University)
Karthik Narasimhan (Princeton University)
Tom Griffiths (Princeton University)

More from the Same Authors