Pre-Learning Environment Representations for Data-Efficient Neural Instruction Following

07/23/2019
by   David Gaddy, et al.
2

We consider the problem of learning to map from natural language instructions to state transitions (actions) in a data-efficient manner. Our method takes inspiration from the idea that it should be easier to ground language to concepts that have already been formed through pre-linguistic observation. We augment a baseline instruction-following learner with an initial environment-learning phase that uses observations of language-free state transitions to induce a suitable latent representation of actions before processing the instruction-following training data. We show that mapping to pre-learned representations substantially improves performance over systems whose representations are learned from limited instructional data alone.

READ FULL TEXT

Please sign up or login with your details

Forgot password? Click here to reset