yanchaosun
|
0c468084
|
sac transfer implementation; disable action encoder
|
5 年前 |
|
yanchaosun
|
0a1a30d3
|
sac update
|
5 年前 |
|
yanchaosun
|
a9c6105d
|
configs
|
5 年前 |
|
yanchaosun
|
00bb821c
|
fix sac transfer problems
|
5 年前 |
|
yanchaosun
|
e2f0b3ca
|
fix transfer
|
5 年前 |
|
yanchaosun
|
cc9a38ae
|
cloud config with shared encoder
|
5 年前 |
|
yanchaosun
|
2b67d1a6
|
fix crawler config
|
5 年前 |
|
yanchaosun
|
42c0c333
|
fig bug
|
5 年前 |
|
yanchaosun
|
d1f57dec
|
separate value net config
|
5 年前 |
|
yanchaosun
|
6df774ed
|
update: separate model train as an option
|
5 年前 |
|
yanchaosun
|
aa0e896f
|
linear value, no target
|
5 年前 |
|
yanchaosun
|
c48b6429
|
numpy fix, config 3dball
|
5 年前 |
|
yanchaosun
|
8c03c82a
|
use target
|
5 年前 |
|
yanchaosun
|
44312bdb
|
linear policy and linear forward
|
5 年前 |
|
yanchaosun
|
57d3ba64
|
change path
|
5 年前 |
|
yanchaosun
|
42c9ba43
|
reuse encoder and linear
|
5 年前 |
|
yanchaosun
|
e8fcc4bb
|
ppo new implementation
|
5 年前 |
|
yanchaosun
|
66bbdae9
|
sac crawler configs
|
5 年前 |
|
yanchaosun
|
120d1c3a
|
cloud config: non-linear policy
|
5 年前 |
|
yanchaosun
|
f78940c1
|
less features
|
5 年前 |
|
yanchaosun
|
3ce88589
|
1 layer everything
|
5 年前 |
|
yanchaosun
|
86da272d
|
load pv
|
5 年前 |
|
yanchaosun
|
6220f7c7
|
linear model
|
5 年前 |
|
yanchaosun
|
f1346bdf
|
multiple seeds
|
5 年前 |
|
yanchaosun
|
de4870be
|
new configs
|
5 年前 |
|
yanchaosun
|
4f64d0f5
|
new config
|
5 年前 |
|
yanchaosun
|
0646e095
|
crawler configs
|
5 年前 |
|
yanchaosun
|
6b8a6e45
|
fix path
|
5 年前 |
|
yanchaosun
|
990d25e3
|
fix path again
|
5 年前 |
|
yanchaosun
|
09e1f0c4
|
another fix
|
5 年前 |
|
yanchaosun
|
15b2e80e
|
action encoder
|
5 年前 |
|
yanchaosun
|
b5e02978
|
sac crawler config
|
5 年前 |
|
yanchaosun
|
5ed6bd3e
|
sac crawler
|
5 年前 |
|
yanchaosun
|
d6f8995a
|
larger feature size
|
5 年前 |
|
yanchaosun
|
ee48cca4
|
linear v
|
5 年前 |
|
yanchaosun
|
49d6b70c
|
crawler: max episode length=1000; new config: 1 forward layer
|
5 年前 |
|
yanchaosun
|
4b081de4
|
smaller feature size
|
5 年前 |
|
yanchaosun
|
96b5478f
|
smaller
|
5 年前 |
|
yanchaosun
|
0463bfe9
|
smaller state feature, large action feature
|
5 年前 |
|
yanchaosun
|
2e927257
|
separate policy net
|
5 年前 |
|
yanchaosun
|
86830ac9
|
3dball mass=5 transfer test
|
5 年前 |
|
yanchaosun
|
dd0ac8a3
|
mass=2
|
5 年前 |
|
yanchaosun
|
46817bed
|
fix bug
|
5 年前 |
|
yanchaosun
|
b0f6f307
|
transfer from mass 2 to mass 1
|
5 年前 |
|
yanchaosun
|
bcdc0a11
|
f512
|
5 年前 |
|
yanchaosun
|
4a23dbb3
|
fix mass 3dball
|
5 年前 |
|
yanchaosun
|
db30f918
|
push block
|
5 年前 |
|
yanchaosun
|
4be4f1d1
|
new reacher env
|
5 年前 |
|
yanchaosun
|
e9a3ea57
|
reacher self-transfer
|
5 年前 |
|
yanchaosun
|
f1802c3a
|
push block transfer setting
|
5 年前 |
|
yanchaosun
|
5cab2114
|
push block without action encoder
|
5 年前 |
|
yanchaosun
|
4133fb35
|
no action
|
5 年前 |
|
yanchaosun
|
191a1133
|
block forward 2 layers
|
5 年前 |
|
yanchaosun
|
1ee62100
|
reacher
|
5 年前 |
|
yanchaosun
|
5c3306ef
|
large buffer size
|
5 年前 |
|
yanchaosun
|
4d5f5888
|
encoder layer 1
|
5 年前 |
|
yanchaosun
|
e39986ed
|
block larger feature size; reacher fix and new reward
|
5 年前 |
|
yanchaosun
|
7dac3284
|
push block more steps
|
5 年前 |
|
yanchaosun
|
51491a3e
|
new dynamics change: scale 1 to 2
|
5 年前 |
|
yanchaosun
|
a1859fb8
|
reacher multi seeds
|
5 年前 |
|
yanchaosun
|
854e10e1
|
3dball hard scale
|
5 年前 |
|
yanchaosun
|
b5a1b9b4
|
hard task name change
|
5 年前 |
|
yanchaosun
|
27dffa4d
|
new reacher reward
|
5 年前 |
|
yanchaosun
|
16e63cb8
|
config fix
|
5 年前 |
|
yanchaosun
|
883361ee
|
reacher new reward: action penalty and constant not-reaching-goal penalty
|
5 年前 |
|
yanchaosun
|
85549b2b
|
reacher: stack observation. with the original reward function
|
5 年前 |
|
yanchaosun
|
92c3facf
|
distance based penalty
|
5 年前 |
|
yanchaosun
|
f15a4f2d
|
2 layers
|
5 年前 |
|
yanchaosun
|
716336bf
|
larger feature size
|
5 年前 |
|
yanchaosun
|
63cec035
|
fix config
|
5 年前 |
|
yanchaosun
|
693c0ca4
|
feature size 32
|
5 年前 |
|
yanchaosun
|
1a9aaaf6
|
model weights and large transfer learning weight
|
5 年前 |
|
yanchaosun
|
1ebe7054
|
new config
|
5 年前 |
|
yanchaosun
|
8f67cd40
|
smaller learning rate
|
5 年前 |
