Latest posts tagged with #offlineRL on Bluesky
Trajectory Data Enables Efficient Offline RL Policy Evaluation
Full trajectory data enables efficient offline RL policy evaluation under linear realizability and concentrability; tighter analysis cuts needed trajectories. Submitted 3 Oct 2025. getnews.me/trajectory-data-enables-... #offlinerl #trajectorydata
One-Step Flow Q-Learning Boosts Offline RL Performance
One-Step Flow Q-Learning (OFQL) uses a single forward pass instead of multi-step denoising, halving inference time and beating Diffusion Q-Learning on the D4RL benchmark. Read more: getnews.me/one-step-flow-q-learning... #offlinerl #diffusion
DiSA‑IQL Enhances Offline Reinforcement Learning for Soft Robot Control
DiSA‑IQL adds a robustness term to Implicit Q‑Learning, penalizing unreliable state‑action pairs; in simulated tests it outperformed BC, CQL and IQL with higher success. Read more: getnews.me/disa-iql-enhances-offlin... #offlinerl #softrobots
In-Context Compositional Q-Learning Boosts Offline RL
ICQL applies linear Transformers to infer local Q‑functions, yielding up to 16.4% higher returns on kitchen tasks and 6.3%–8.6% gains on Gym and Adroit benchmarks. Read more: getnews.me/in-context-compositional... #offlinerl #transformers #reinforcementlearning
Offline RL Boosts Multi-Agent Path Finding via GPT-4o
Researchers combined offline reinforcement learning with GPT‑4o to speed up multi‑agent path finding, cutting training from weeks to hours and boosting success while cutting collisions. Read more: getnews.me/offline-rl-boosts-multi-... #offlinerl #gpt4o #mapf
DAWM Model Boosts Offline Reinforcement Learning Diffusion Actions
DAWM splits generation: a diffusion model predicts future states and rewards, while an inverse dynamics model infers actions. DAWM‑augmented data boosted TD3BC and IQL on D4RL. Read more: getnews.me/dawm-model-boosts-offlin... #offlinerl #diffusion
Offline RL Improves Job-Shop Scheduling Using Limited Data
Researchers unveiled CDQAC, an offline RL algorithm that learns job‑shop scheduling from only 10‑20 historic instances and beats the heuristics and RL baselines. (12 Sep 2025) Read more: getnews.me/offline-rl-improves-job-... #offlineRL #jobscheduling
Hybrid Adaptive Conformal Offline RL Improves Fair Medicaid Care
HACO achieved an AUC of ~0.81 and set a risk threshold τ≈0.038 (α=0.10) to safely guide Medicaid population health decisions. Read more: getnews.me/hybrid-adaptive-conforma... #offlinerl #medicaid
