Mirrors_Framework/deepmind-research
1
0
Fork 0
mirror of https://github.com/google-deepmind/deepmind-research.git synced 2026-05-10 13:27:17 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

Fix file name in README for gamma=1 images.

Browse Source 
PiperOrigin-RevId: 294215051
This commit is contained in:
Josh Abramson
2020-02-10 15:31:27 +00:00
committed by Diego de Las Casas
parent 48ba42f792
commit d0efbec03a
1 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tvt/README.md
+2 -2
View File
@@ -145,7 +145,7 @@ low performance.<br>
For 10 replicas without TVT and with gamma equal to 1, performance of the RMA
agent without TVT is improved, but is unstable and never consistently goes above
6.<br>
# ![No_TVT_ktd_gamma1](images/ktd_no_tvt_gamma1.png)
# ![No_TVT_ktd_gamma1](images/ktd_notvt_gamma1.png)
### Active-visual-match
Across 10 replicas, we found that the TVT agents get to a score of 10,
@@ -162,7 +162,7 @@ For 10 replicas wihtout TVT and with gamma equal to 1, performance of the RMA
agent without TVT
is considerably worse, suggesting the behavior learnt from later phases does not
result in undirected exploration in the first phase.
# ![No_TVT_vm_gamma_1](images/avm_no_tvt_gamma1.png)
# ![No_TVT_vm_gamma_1](images/avm_notvt_gamma1.png)
## Citing this work