eloialonso/iris
Transformers are Sample-Efficient World Models. ICLR 2023, notable top 5%.
PythonGPL-3.0
Issues
- 2
IRIS agent in Hugging face transformers
#26 opened by RUFFY-369 - 1
- 4
please gab code (delta-IRIS )
#20 opened by wassname - 1
- 1
World model working
#24 opened by suryadheeshjith - 1
How to reproduce the results of IRIS?
#21 opened by qiwang067 - 1
predict the first token of next obs?
#19 opened by abliao - 3
The role of sampling_weights.
#10 opened by Yingdong-Hu - 1
sample_weights problem
#18 opened by abliao - 2
Compatibility of continuous action space
#16 opened by dibbla - 2
About the study restart function
#17 opened by tatematsu123 - 1
- 3
update to continuous actions
#13 opened by 2M-kotb - 1
Actor Critic training time
#12 opened by Recoan0 - 2
Size mismatch when evluating world model
#14 opened by mamengyiyi - 1
Question about the training time
#11 opened by zzzc18 - 1
Training in multi-GPU system
#9 opened by lyp741 - 4
- 2
About world model training
#8 opened by rook86 - 2
- 1
MSE for reward predictor
#5 opened by MikeTkachuk - 3
ForkingPickler: Can't pickle local object
#4 opened by hexaflexa - 1
System Requirements
#2 opened by robjlyons