Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Offline Reinforcement Learning on Antmaze Medium play offline (target domain)

398.8Target Domain Score (Normalized)

LoDADA

-2.64101.58205.8310.02Feb 24, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.02
398.8
2026.02
378.8
2026.02
360.4
2026.02
352.4
2026.02
288.4
2026.02
158.4
2026.02
76.4
2026.02
74.8
2026.02
74.4
2026.02
73.6
2026.02
71.6
2026.02
71.6
2026.02
70.8
2026.02
69.6
2026.02
67.6
2026.02
66.8
2026.02
66.4
2026.02
64.8
2026.02
63.2
2026.02
62.8
2026.02
60.8
2026.02
58
2026.02
58
2026.02
56.8
2026.02
56.4
2026.02
55.2
2026.02
54.4
2026.02
52.8
2026.02
52.8
2026.02
52.4
2026.02
48.8
2026.02
46.8
2026.02
44.8
2026.02
42.4
2026.02
34.4
2026.02
34
2026.02
33.6
2026.02
30
2026.02
29.6
2026.02
25.2
2026.02
22.8
2026.02
12.8