issues
search
vlad17
/
mve
MVE: model-based value estimation
Apache License 2.0
10
stars
0
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
rm reward scaling
#358
vlad17
closed
6 years ago
0
Add humanoid
#357
vlad17
closed
6 years ago
0
reporting separate from timestep advancing
#356
vlad17
closed
6 years ago
0
Optional dataset persist
#355
vlad17
closed
6 years ago
0
bbox legend plotting
#354
vlad17
closed
6 years ago
0
Ddpg plot
#353
alvinwan
closed
6 years ago
0
change font scale in main_evaluate_qval
#352
vlad17
closed
6 years ago
0
Accelerate MVE DDPG
#351
vlad17
closed
6 years ago
1
document parallelism and allow for finer grained TF parallelism control
#350
vlad17
closed
6 years ago
0
increase font size
#349
vlad17
closed
6 years ago
0
Additional regularization strategies for dynamics nn
#348
vlad17
closed
6 years ago
0
use explicit boolean flags in flags that get tuned
#347
vlad17
closed
6 years ago
0
misc convenience bugs in ray launch scripts
#346
vlad17
closed
6 years ago
0
add batch norm to dynamics
#345
vlad17
closed
6 years ago
0
add additional dynamics reporting
#344
vlad17
closed
6 years ago
0
added parallelism flag
#343
vlad17
closed
6 years ago
0
update docs
#342
vlad17
closed
6 years ago
0
extract infra into separate packages
#341
vlad17
closed
6 years ago
1
fix threading
#340
vlad17
closed
6 years ago
0
migrate to ray 0.3.1
#339
vlad17
closed
6 years ago
0
put main files in separate directory
#338
vlad17
closed
6 years ago
1
Advance correctly (sample first, then report, then train)
#337
vlad17
closed
6 years ago
0
get rid of experiment_main
#336
vlad17
closed
6 years ago
0
convert to absl flags
#335
vlad17
closed
6 years ago
1
counter now correctly skips over skipped evaluations
#334
vlad17
closed
6 years ago
0
move "should_*" logic in experiment to some kind of online counter class
#333
vlad17
closed
6 years ago
0
Q-value density plot
#332
vlad17
closed
6 years ago
0
added title arg
#331
alvinwan
closed
6 years ago
3
*_every overzealous reporting
#330
vlad17
closed
6 years ago
1
convert to timesteps-based iteration
#329
vlad17
closed
6 years ago
0
imaginary buffer
#328
vlad17
closed
6 years ago
0
move omp_num_threads dependence to a flag dependence
#327
vlad17
closed
6 years ago
0
sparser reporting
#326
vlad17
closed
6 years ago
0
Friendly norm
#325
vlad17
closed
6 years ago
0
track dynamics normalization stats
#324
vlad17
closed
6 years ago
1
Classical control envs
#323
vlad17
opened
6 years ago
0
made TD-k optional
#322
vlad17
closed
6 years ago
0
replace tf.AUTO_REUSE with explicit reuse patterns
#321
vlad17
opened
6 years ago
0
Thread fix
#320
vlad17
closed
6 years ago
0
remove Q mse bias recording in ddpg
#319
vlad17
closed
6 years ago
0
sensible reporting names for dynamics metrics
#318
vlad17
closed
6 years ago
1
tf_action -> tf_target_action in critic expansion
#317
vlad17
closed
6 years ago
0
Accelerate ddpg w/ learned dyn
#316
vlad17
closed
6 years ago
0
profile and speed up ddpg with learned dynamics
#315
vlad17
closed
6 years ago
0
consistent units
#314
vlad17
closed
6 years ago
1
consistent units
#313
vlad17
closed
6 years ago
0
Learned dynamics TD-k fix
#312
vlad17
closed
6 years ago
0
Pusher
#311
alvinwan
closed
6 years ago
2
Hopper
#310
alvinwan
closed
6 years ago
0
Sampler fixes
#309
vlad17
closed
6 years ago
2
Previous
Next