issues
search
jhejna
/
cpl
Code for Contrastive Preference Learning (CPL)
MIT License
147
stars
12
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Why did you use a ContinuousActor?
#14
CAI23sbP
closed
2 months ago
2
Problem aboult the dimension of the observation space
#13
zyw19970608
closed
4 months ago
0
The purpose of `contrastive_bias`
#12
pengzhenghao
closed
2 months ago
2
Evaluation Procedure
#11
hang-wu
closed
7 months ago
2
Combining CPL and action chunking does work! Would you like to slightly modify the codebase to an action chunk version?
#10
StarCycle
opened
7 months ago
1
Upload the doc version of the proof
#9
StarCycle
opened
7 months ago
0
Nice work!
#8
pengzhenghao
closed
7 months ago
1
Add explanation of the biased BCE loss used
#7
StarCycle
closed
7 months ago
2
Difference between 'reward' and 'advantage' in p-iql?
#6
DooHyun-Lee
closed
7 months ago
2
How to create datasets?
#5
gyh-ustc
closed
8 months ago
2
[Bug?]: Questions about the TD target calculation in P-IQL
#4
typoverflow
closed
8 months ago
2
pip install -e research error.
#3
yiqiaoqingyuyu
closed
8 months ago
2
Improved Readability
#2
Sanyam-2026
closed
2 months ago
0
Fix typo in README.md
#1
eltociear
closed
11 months ago
1