method
active
method:hard-parameter-sharing-hpsHard-parameter sharing (HPS)
Architecture pattern with a shared encoder and task-specific heads.
Neighborhood — ranked by edge-count
Artifacts (1)
artifact
- The paper proposing the Dual-Balancing Multi-Task Learning method.
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- Parameter shared across all tasks, typically the encoder.
- Demonstrates architecture-agnostic applicability of the SAE tuning method
- Stress sharing is an easy way to achieve robustness in collectives made up of homeostatic subunits.claim0.692Generalization of the model's implications.
- Stress-sharing populations reach anatomical targets faster than hardwired or non-sharing populations.finding0.690Populations with stress sharing discovered correct morphology by generation 500, vs non-sharing and hardwired (p≪0.01).
- Exhaustive search over 312,130 subjective reward functions per environment to find best-performing agents
- The Hard Problem is likely unsolvable in 3rd person and should not stand as a barrier to progress.claim0.673Position that the Hard Problem should not halt applied sentience assessment.
- Demonstrates benefit of stress sharing across smaller grid complexity.
- Parameter specific to each task, e.g., task head.