Coverage figure: sixteen cells, all learn - Gaussian square (E40), fpo-policy · FPO · square (E41), square plots success by tactino · Pull Request #97 · PlugRL/plugrl-server

exp: E41 scripts - E39 resumed to FPO++'s 8M steps (before the pilot)

tactino

committed

af3fc82

exp: E41 run_cell.sh - no apostrophe inside ${E39:?...}

tactino

committed

1127d39

exp: pre-register E41 - E39 resumed to FPO++'s 8M steps, the extension declared as decided after E39

tactino

committed

3898cce

Merge remote-tracking branch 'origin/exp/e40-square-gaussian-ppo' into figure/square-gaussian-pi0

tactino

committed

d0c2ae2

Merge remote-tracking branch 'origin/exp/e36-pi0-longer' into figure/square-gaussian-pi0

tactino

committed

7b363b9

figure: the Gaussian MLP row's square cell from E40, and pi0.5's cells from E36

tactino

committed

2467d09

Merge remote-tracking branch 'origin/main' into figure/square-gaussian-pi0

tactino

committed

54e5edf

README: the Gaussian MLP row is CleanRL's on MuJoCo and DPPO's on square

tactino

committed

3c2c054

coverage: resumed_from - a cell whose runs resumed from earlier ones plots the whole run

tactino

committed

a443cb8

coverage: fpo-policy · FPO · square learns - E39 resumed to FPO++'s 8M steps by E41, two of three seeds past the bar

tactino

committed

b511355

Merge remote-tracking branch 'origin/exp/e41-square-fpo-plus-plus-8m' into figure/square-gaussian-pi0

tactino

committed

6a9ec14

exp: E41 results - E39 resumed to FPO++'s 8M steps learns square by the bar, two of three seeds past it, all three down from their best

tactino

committed

156d0ef

Merge branch 'exp/e41-square-fpo-plus-plus-8m' into figure/square-gaussian-pi0

tactino

committed

9170521

README: sixteen cells, every pair learns every task

tactino

committed

71026d2

coverage: square plots the success rate - its runs do not share a reward

tactino

committed

63827d3

coverage: pi0-policy · DPPO note - 20 of 50 is on the holding line and below the untrained 31, not 'barely moved'

tactino

committed

39bceca

README: pi0.5 on LIBERO has a page of its own; the grid is the home page's figure

tactino

committed

e64b1a2

Navigation Menu

Coverage figure: sixteen cells, all learn - Gaussian square (E40), fpo-policy · FPO · square (E41), square plots success - #97

Coverage figure: sixteen cells, all learn - Gaussian square (E40), fpo-policy · FPO · square (E41), square plots success#97
tactino merged 17 commits into
mainfrom
figure/square-gaussian-pi0

Commits on Sep 28, 2026

exp: E41 scripts - E39 resumed to FPO++'s 8M steps (before the pilot)

exp: E41 run_cell.sh - no apostrophe inside ${E39:?...}

exp: pre-register E41 - E39 resumed to FPO++'s 8M steps, the extension declared as decided after E39

Merge remote-tracking branch 'origin/exp/e40-square-gaussian-ppo' into figure/square-gaussian-pi0

Merge remote-tracking branch 'origin/exp/e36-pi0-longer' into figure/square-gaussian-pi0

figure: the Gaussian MLP row's square cell from E40, and pi0.5's cells from E36

Merge remote-tracking branch 'origin/main' into figure/square-gaussian-pi0

README: the Gaussian MLP row is CleanRL's on MuJoCo and DPPO's on square

coverage: resumed_from - a cell whose runs resumed from earlier ones plots the whole run

coverage: fpo-policy · FPO · square learns - E39 resumed to FPO++'s 8M steps by E41, two of three seeds past the bar

Merge remote-tracking branch 'origin/exp/e41-square-fpo-plus-plus-8m' into figure/square-gaussian-pi0

exp: E41 results - E39 resumed to FPO++'s 8M steps learns square by the bar, two of three seeds past it, all three down from their best

Merge branch 'exp/e41-square-fpo-plus-plus-8m' into figure/square-gaussian-pi0

README: sixteen cells, every pair learns every task

coverage: square plots the success rate - its runs do not share a reward

coverage: pi0-policy · DPPO note - 20 of 50 is on the holding line and below the untrained 31, not 'barely moved'

README: pi0.5 on LIBERO has a page of its own; the grid is the home page's figure