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

HT Trigger threshold control on Monte Carlo (MC) (held-out 20%)

0.004MAE

GFPO-F

Updated 1mo ago

Evaluation Results

Method	Links
GFPO-F 2026.06		0.004	0.01	100	99.425	34.273	99.425	34.273
L-GRPO 2026.06		0.006	0.016	100	99.425	34.293	99.425	34.293
DQN 2026.06		0.01	0.023	100	99.423	34.255	99.423	34.255
ADT 2026.06		0.01	0.023	100	99.423	34.255	99.423	34.255
CPO 2026.06		0.01	0.023	100	99.423	34.255	99.423	34.255
PPO 2026.06		0.013	0.029	88.9	99.439	34.544	99.434	34.492
GFPO-FR 2026.06		0.016	0.024	100	99.468	35.057	99.468	35.057
DQN-F 2026.06		0.018	0.038	74.3	99.409	34.141	99.417	34.184
GRPO 2026.06		0.024	0.052	63.2	99.422	34.396	99.401	34.106
PID 2026.06		0.029	0.07	52.1	99.388	33.289	99.294	31.444
DSPOT 2026.06		0.031	0.088	60.4	99.355	32.612	99.3	31.718
Constant 2026.06		0.083	0.191	25	98.989	28.062	99.129	29.418