Skip to content

Official code for paper: Conservative objective models are a special kind of contrastive divergence-based energy model

Notifications You must be signed in to change notification settings

christopher-beckham/coms-are-energy-models

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 
 
 
 
 

Repository files navigation

coms-are-energy-models

Official code for the paper: Conservative objective models are a special kind of contrastive divergence-based energy model

In this work we theoretically show that conservative objective models (COMs) for offline model-based optimisation (MBO) are a special kind of contrastive divergence-based energy model, one where the energy function represents both the unconditional probability of the input and the conditional probability of the reward variable. While the initial formulation only samples modes from its learned distribution, we propose a simple fix that replaces its gradient ascent sampler with a Langevin MCMC sampler. This gives rise to a special probabilistic model where the probability of sampling an input is proportional to its predicted reward. Lastly, we show that better samples can be obtained if the model is decoupled so that the unconditional and conditional probabilities are modelled separately.

./assets/coms-animation-compressed.gif

Reproduction

All experiments are self-contained in the IPython notebook file!