Anton Bakhtin, Eltayeb Ahmed, Laurens Van Der Maaten, Rohit Girdhar
We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. The repositories linked to it are listed below.
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
A common approach to solving physicalreasoning tasks is to train a value learner on example tasks. A limitation of such an approach is that it requires learning about object dynamics solely from reward values assigned to the final state of a rollout of the environment. This study aims to address this limitation by augmenting the reward value with self-supervised signals about object dynamics. Specifically, we train the model to characterize the similarity of two environment rollouts, jointly with predicting the outcome of the reasoning task. This similarity can be defined as a distance measure between the trajectory of objects in the two rollouts, or learned directly from pixels using a contrastive formulation. Empirically, we find that this approach leads to substantial performance improvements on the PHYRE benchmark for physical reasoning (Bakhtin et al., 2019), establishing a new state-of-the-art.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2102.10336")
get_code_for_paper("2102.10336")
have("2102.10336")
The run record, dated, one paper per request, free:
curl https://syntology.ai/api/ran/2102.10336.json
A badge for a README (the split and the date, never a ratio):
[](https://syntology.ai/paper/2102.10336)
Connect an agent — have() is free.