TellWell
← Back to feed
Publications3h ago88% confidenceConfidence 88% — the share of independent, credible sources corroborating the core facts.

DoorDash Deploys Multi-Agent Reinforcement Learning to Optimize Food Delivery Dispatch

Center 100%
1 source

DoorDash researchers have developed and deployed a reinforcement learning system that automatically adjusts dispatch priorities in its food delivery marketplace using real operational feedback. The system learns from delayed signals like delivery speed and courier utilization to optimize the tradeoff between delivery quality and batching efficiency. This approach demonstrates how machine learning can safely improve logistics decisions in large-scale real-world systems while maintaining operational constraints.

Researchers at DoorDash have published details of a deployed reinforcement learning system designed to optimize dispatch decisions in three-sided food delivery marketplaces. Rather than replacing the existing combinatorial assignment optimizer, the system learns a store-level policy that selects discrete multipliers to adjust the optimizer's tradeoff between delivery quality and batching efficiency. The approach uses centralized offline learning from logged marketplace data combined with decentralized execution, employing Double Q-learning with conservative regularization to handle noisy, delayed, and coupled feedback signals. A production switchback experiment showed the offline-trained policy increased batching efficiency and reduced courier time costs without degrading customer-facing delivery quality. The work illustrates how reinforcement learning can be safely integrated into complex real-world economic and logistics systems while preserving production feasibility and operational safeguards.

What different sources said

  • Multi-Agent Reinforcement Learning from Delayed Marketplace Feedback for Objective-Weight Adaptation in Three-Sided Dispatch

Related

PublicationsConfidence 78% — the share of independent, credible sources corroborating the core facts.

Topology-Aware Thermodynamics Improves DNA Probe Specificity Design

Researchers developed a new framework for designing DNA probes that accounts for the spatial organization of matched sequences, not just overall thermodynamic stability. Traditional methods rely on scalar measures like melting temperature and free energy, which miss how mismatches are distributed along the probe. The approach could improve diagnostic accuracy in applications like HPV detection and gene expression profiling.

1 source2h ago
PublicationsConfidence 82% — the share of independent, credible sources corroborating the core facts.

Study Identifies Optimal Thermal Dose for Combining Focused Ultrasound with Immunotherapy in Tumors

Researchers used multimodal PET imaging to identify an optimal thermal dose range for focused ultrasound ablation that destroys tumor tissue while preserving conditions for immunotherapy delivery. The study found that excessive heating collapses blood vessels needed for antibody access, while insufficient heating fails to adequately reduce tumor burden. The findings could guide clinical design of combination treatments pairing thermal ablation with immunotherapies.

1 source3h ago
PublicationsConfidence 88% — the share of independent, credible sources corroborating the core facts.

Plant MSH1 Protein Functions as Mismatch-Directed Nuclease for Organelle Genome Maintenance

Researchers have identified the precise mechanism by which the AtMSH1 protein in Arabidopsis plants recognizes and cleaves DNA mismatches and lesions, preventing mutations in organellar genomes. The protein combines a DNA mismatch recognition module with a nuclease domain that makes staggered cuts at specific positions relative to DNA damage. This discovery explains how plants maintain unusually low mutation rates in their mitochondrial and chloroplast DNA compared to other eukaryotes.

1 source3h ago