10 citations · 22 across the 10 of their papers we have counts for
Showing cs.LGShow all
2 papers · 1 filter
cs.LG2026
No Gaussian Required: Contrastive Inverse Dynamics for JEPA World Models
Jack Boylan, Chris Hokamp
Joint-Embedding Predictive Architectures (JEPAs) learn world models by predicting future embeddings, but the objective admits a trivial solution of a constant encoder, so every pra…
cs.LG2019★ 7 cited
Task Selection Policies for Multitask Learning
John Glover, Chris Hokamp
One of the questions that arises when designing models that learn to solve multiple tasks simultaneously is how much of the available training budget should be devoted to each indi…