Mean final packing density per policy on the training distribution
and five unseen distributions: the EMS heuristic baseline, our learned
policy, and the full-sequence packing oracle.
Distribution
EMS
Our Policy
Oracle
Ours / oracle
Figure 3. Our policy is averaged over 200 rollouts per distribution (20
sequences × 10 seeds); EMS and the oracle over 20 (20 sequences × 1 seed,
policies are deterministic).
Error bars are ±1 standard deviation. Note that the y axis is truncated at
40% rather than starting at zero, so bar lengths are not proportional to
density. Hover or focus the chart and use
arrow keys for exact values; the table view lists them all.