lasalaai.bsky.social
👤 Joined on 02 September 2025
Followers
Following
Total Posts
@lasalaai.bsky.social🔊Simplicial Embeddings (SEMs) Improve Sample Efficiency in Actor-Critic Agents🔊 In our recent preprint we demonstrate that the use of well-structured representations (SEMs) can dramatically improve sample efficiency in RL agents. 1/X
Show PostWe first show that non-stationarity amplifies representation collapse by evaluating on a CIFAR-10 experiment where we create non-stationarity by randomly shuffling labels every 20 epochs. Using SEMs helps avoid collapse, reduces neuron dormancy, and results in lower loss. 3/X
Show PostWe use FastTD3 (arxiv.org/abs/2505.22642), a recent baseline for continuous control which is performant and fast, to evaluate the benefits of SEM. As seen in the second figure, SEMs yield improvements throughout. 4/X
Show PostWhy do SEMs improve performance? Our analyses show they increase effective rank of actor features while bounding their norms, have reduced losses, more consistency across critics, & sparser representations, ultimately resulting in improved performance and sample efficiency. 5/X
Show PostWe find SEM to be the most effective when compared against alternative approaches for structuring representations. We also evaluated the impact of the choices of V and L, finding that higher V with lower L seems to yield the best results. 6/X
Show Post