-
Hi~Thank you very much for sharing your paper and source code !!! I am new to inverse RL and I want to implement your method on the robot recently.
**About Ant-v2**
1. And I found that the reward fo…
-
Below are some algorithms that it would be nice to see in `imitation`, but which aren't urgently needed. Feel free to extend this list.
Learning from demonstrations:
- [ ] [IQ-Learn](https://githu…
-
- Supprimer la portée
- Remanier le champs de tags afin de proposer des tags reliés à la description
- Rendre le courriel personne qui soumet facultatif et ajouter une mention comme quoi cela est fac…
-
## Problem
Robotic Env such as [SurRoL]( https://github.com/med-air/SurRoL) and [Fetch](https://robotics.farama.org/envs/fetch/) uses Dictionary Observation Space, with
1. observation
2. desired_g…
-
The default designs in SR2 have never been particularly impressive, to the point that I don't think anyone uses their starter ships for anything except scouting (unless they retrofit them, but this op…
-
## Bug description
```
> assert reward_improvement.is_significant_reward_improvement(
rewards_before, # type:ignore[arg-type]
rewards_after, # type:ignore[arg-type]…
-
**Is your feature request related to a problem? Please describe.**
A clear and concise description of what the problem is. Ex. I'm always frustrated when [...]
BC algorithm can fall victim to covari…
-
## Problem
There is currently no script in `src/imitation/scripts` that supports training SQIL implemented in `src/imitation/algorithms/sqil.py`. This both makes the algorithm harder to use, and is i…
-
## Problem
The current implementation of SQIL (PR: #744 , issue: #740 ) only works with DQN while the original paper also used soft-actor-critic and soft Q-learning.
It would be great to have sup…
-
Hi, thanks for your wonderful paper. I have been also trying with absorbing states in IQ-learn also but I don't know why it works well.
**My question is that is it fine to update policy using absorb…