mila-iqia/atari-representation-learning
Code for "Unsupervised State Representation Learning in Atari"
PythonMIT
Pinned issues
Issues
- 0
Which version of gym[atari] was used?
#74 opened by Sopralapanca - 0
Conflict to the requirement
#72 opened by hyyh28 - 0
- 5
Can't reproduce experimental results
#69 opened by Yingdong-Hu - 0
- 1
How necessary is local-local infomax?
#66 opened by seann999 - 2
Generating Enough Episodes for Tennis
#65 opened by adamtupper - 3
Ram annotation of (x, y) are not aligned
#64 opened by happywu - 2
unused naff_fc_size parameter
#63 opened by AnxietyYoungPoet - 3
Incorrect/ambiguous features in Seaquest
#59 opened by damnOblivious - 0
Nearest neighbor visualizations
#1 opened by ankeshanand - 0
Add actions to dataloader
#7 opened by eracah - 0
Inverse Model
#8 opened by eracah - 4
Intuition on cross entropy.
#58 opened by biggzlar - 3
Benchmark the pretrained-rl-agent method
#55 opened by liuyuezhang - 9
Pillow version problem
#52 opened by DuaneNielsen - 1
- 2
- 2
Evaluate "transporter" network
#51 opened by DuaneNielsen - 2
Pixel positions
#47 opened by frederikschubert - 1
Overwriting of src
#44 opened by eracah - 5
Any tips on extracting RAM locations?
#40 opened by neighthan - 6
Reproducing published score
#41 opened by htdt - 1
Pretrained PPO trajectories
#39 opened by bmazoure - 1
Figure out a better way to collect samples to train the encoder (other than a random policy)
#13 opened by ankeshanand - 0
- 1
CPC implementation
#2 opened by ankeshanand - 0
Control: Evaluate how well the pretrained representations perform on control tasks
#12 opened by ankeshanand - 1
- 1
Add VAE baseline
#24 opened by eracah - 1
- 1
add atari uber rollouts to run_contrastive code
#33 opened by eracah - 1
- 1
- 1
Spatio-Temporal Estimator
#25 opened by ankeshanand - 1
Make encoders work for (210, 160) images
#31 opened by ankeshanand - 1
Random CNN probe baseline
#26 opened by eracah - 2
- 1
Add validation evaluation to contrastive training
#23 opened by eracah - 2
- 1
- 3
Fix bucketing issues
#15 opened by eracah - 1
Add validation set for probing
#14 opened by ankeshanand - 1
- 1
Design document for the probing benchmark
#10 opened by ankeshanand - 1
Setup the linear probe pipeline
#3 opened by ankeshanand - 1