Nicolas Espinosa Dice
@nico-espinosa-dice.bsky.social
cs phd student @cornelluniversity.bsky.social. previously @harveymuddcollege.bsky.social. working on reinforcement learning & generative models. https://nico-espinosadice.github.io/
Reposted by Nicolas Espinosa Dice
Shortcut models enable scaling offline RL, both at train-time at test-time! We beat so many other algorithms on so many tasks we had to stick most of the results in the appendix 😅. Very proud of @nico-espinosa-dice.bsky.social for spearheading this project, check out his thread!
by incorporating self-consistency during offline RL training, we unlock three orthogonal directions of scaling:
1. efficient training (i.e. limit backprop through time)
2. expressive model classes (e.g. flow matching)
3. inference-time scaling (sequential and parallel)
1. efficient training (i.e. limit backprop through time)
2. expressive model classes (e.g. flow matching)
3. inference-time scaling (sequential and parallel)
June 12, 2025 at 11:14 PM
Shortcut models enable scaling offline RL, both at train-time at test-time! We beat so many other algorithms on so many tasks we had to stick most of the results in the appendix 😅. Very proud of @nico-espinosa-dice.bsky.social for spearheading this project, check out his thread!
by incorporating self-consistency during offline RL training, we unlock three orthogonal directions of scaling:
1. efficient training (i.e. limit backprop through time)
2. expressive model classes (e.g. flow matching)
3. inference-time scaling (sequential and parallel)
1. efficient training (i.e. limit backprop through time)
2. expressive model classes (e.g. flow matching)
3. inference-time scaling (sequential and parallel)
June 12, 2025 at 7:34 PM
by incorporating self-consistency during offline RL training, we unlock three orthogonal directions of scaling:
1. efficient training (i.e. limit backprop through time)
2. expressive model classes (e.g. flow matching)
3. inference-time scaling (sequential and parallel)
1. efficient training (i.e. limit backprop through time)
2. expressive model classes (e.g. flow matching)
3. inference-time scaling (sequential and parallel)
I think of misspecification (e.g. embodiment / sensory gaps) as the fundamental reason behavioral cloning isn't "all you need" for imitation as matching actions matching outcomes. Introducing @nico-espinosa-dice.bsky.social's #ICLR2025 paper proving that "local search" *is* all you need! [1/n]
April 7, 2025 at 10:08 PM