Driving · 2026-10-02
Cross-entropy optimization with prioritized constraints
Francisco Roldan Sanchez, Pau de las Heras Molins, David Fridovich-Keil, Georgios Bakirtzis
AI-assisted summary based on the paper. Consult the linked source for methods and limitations. Benchmark numbers appear after editor verification.
TL;DR
The authors introduce TierCEM, a variant of the cross-entropy method that enforces strict constraint priorities directly in elite selection via sequential filtering of sampled candidates from highest- to lowest-priority constraint. When a tier would eliminate all candidates, TierCEM relaxes using the blocking constraint while preserving all higher-priority constraints. They evaluate on 2D navigation and contact-rich pushing, including learned world-model margins, showing that reversing priority order changes which constraints are violated under conflict.
Why it matters
Many constrained planners collapse priorities into weighted trade-offs that depend on weight calibration. TierCEM instead encodes lexicographic priority in the CEM elite selection logic, so reversing constraint order changes relaxation behavior without requiring per-constraint importance weights. The paper also analyzes how prioritizing task progress enables relaxing lower-priority constraints to avoid stalling.
Method
- TierCEM samples candidate trajectories and applies constraints through an ordered priority-relaxation cascade during elite selection.
- It sequentially filters candidates from highest- to lowest-priority; if a tier empties the set, it falls back to the last nonempty set and ranks by the blocking constraint’s smallest violations, recursively preserving higher tiers.
- If all constraints retain candidates, it selects elites by ranking with the lowest-priority constraint margin (m_N(k)).
Limitation
The stalling mechanism depends on dynamics that admit a stationary constraint-satisfying solution; it does not arise in the inertial setting examined in Appendix E.
Abstract (from arXiv)
When constraints conflict, an optimizer must determine which requirements to preserve and which to relax. On the one hand, a priority ordering specifies which requirements take precedence. On the other hand, penalty-based formulations encode their relative importance through numerical weights. Depending on these weights, a solution can improve its weighted score while violating intended priorities. We introduce TierCEM, a variant of the cross-entropy method that incorporates strict constraint priorities directly into elite selection without requiring per-constraint importance weights. TierCEM works by sequentially filtering sampled candidates, from highest- to lowest-priority constraint. If and when a constraint eliminates all remaining candidates, TierCEM returns to the last nonempty set and selects elites with the smallest violations of that blocking constraint, recursively preserving satisfaction of all higher-priority constraints. We evaluate TierCEM on 2D navigation and contact-rich pushing tasks in proprioceptive and learned world-model settings. Experiments show that reversing the constraint ordering changes which constraints are violated under conflict. Prioritizing progress toward the task objective also enables TierCEM to relax lower-priority constraints when they would otherwise prevent further progress.
Related papers
- EpicWorldModel: Exploration-driven Planning with Latent World Models
- In CEM, a World Model Is Also a Proposal Mechanism
- World-Model Policy Arbiter for Goal-Conditioned Reinforcement Learning
- Knowing When to Stop Thinking: A Dissection of Expected Free Energy as a Test-Time Compute Controller for Latent World-Model Planning
- G$^2$WAM: World Action Models for Efficient and Geometry Grounded Image-Goal Navigation
Publication record
Bibliographic evidence describes where and when a record appeared. It does not validate scientific claims.
First public release: 2026-10-01 · Indexed by Pub-AI: 2026-10-02 UTC
Cross-entropy optimization with prioritized constraints
Preprint posted: 2026-10-01 · Original record ↗
- arxiv
- 2610.01319
Explicit source versions (1)
- v1 · 2026-10-01
- Preprint available
Source date: 2026-10-01 · First detected: 2026-10-25 UTC · Historical baseline
Source evidence (1)
- Official evidence ↗ · atom:published (official metadata)
Abstract · arxiv
When constraints conflict, an optimizer must determine which requirements to preserve and which to relax. On the one hand, a priority ordering specifies which requirements take precedence. On the other hand, penalty-based formulations encode their relative importance through numerical weights. Depending on these weights, a solution can improve its weighted score while violating intended priorities. We introduce TierCEM, a variant of the cross-entropy method that incorporates strict constraint priorities directly into elite selection without requiring per-constraint importance weights. TierCEM works by sequentially filtering sampled candidates, from highest- to lowest-priority constraint. If and when a constraint eliminates all remaining candidates, TierCEM returns to the last nonempty set and selects elites with the smallest violations of that blocking constraint, recursively preserving satisfaction of all higher-priority constraints. We evaluate TierCEM on 2D navigation and contact-rich pushing tasks in proprioceptive and learned world-model settings. Experiments show that reversing the constraint ordering changes which constraints are violated under conflict. Prioritizing progress toward the task objective also enables TierCEM to relax lower-priority constraints when they would otherwise prevent further progress.
