reinforcement learning

Adaptive operator selection via deep reinforcement learning for decomposition-based multi-objective multi-task optimization

Multi-objective multi-task optimization (MO-MTO), which aims to simultaneously solve multiple related multi-objective problems by leveraging knowledge transfer across tasks to enhance optimization performance on each task, has emerged as a new research focus in the field of evolutionary computation. Although a variet...

Multi-axis CNC machining and toolpath strategies for complex free-form geometries: a systematic review of geometry, kinematics, process physics and intelligent optimization

In addition to offering the enabling process for the turbine blade, blisks, impellers, moulds, dies and other free-form components, multi-axis computer numerical control (CNC) machining introduces the geometry of toolpath, kinematics of machine, cutting mechanics, surface integrity and energy use. In this paper, resear...

Memory formation as selection: an eligibility-based architecture

Memory is often discussed in terms of encoding strength, retention, and retrieval success. Yet these dimensions do not fully explain how remembered information acquires structure within a larger mnemonic system. Strongly retained or highly accessible traces need not be broadly integrated, and their persistence or vivid...

Sequential–hierarchical coordination for multi-task enroute air traffic control

Enroute air traffic controllers (ATCOs) face two fundamentally distinct yet interrelated operational tasks: macroscopic flow-level planning and flight level assignment for sector-wide traffic optimization over time horizons on the order of tens of minutes and microscopic real-time conflict detection and resolution (CD&...

Reinforcement learning for sustainable construction scheduling: a many-objective framework with adaptive search and multi-criteria decision support

Delivering construction projects sustainably requires reconciling duration, cost, carbon emissions and the stability and utilization of site resources, yet these objectives conflict and the choice among candidate schedules is seldom made on an explicit, data-driven basis. This study aims to develop a data-driven de...

Radar-vision-V2X fused perception based intelligent control mechanism for service-oriented traffic signals

Traditional traffic signal systems lack the capability to perceive vehicle types and real-time traffic flow dynamics, particularly in identifying priority levels for emergency vehicles. To address this, an intelligent traffic flow decision-making and scheduling system based on radar-visual-V2X(Vehicle-to-Everything) mu...

Scooter-Driving Humanoid Robot: Sim-to-Real Transfer Through Deep Reinforcement Learning

This paper presents a simulation-to-reality (sim-to-real) transfer process, allowing a full-sized humanoid robot to autonomously balance a two-wheeled scooter and track operator-supplied heading commands through a deep reinforcement learning (DRL) policy, including while carrying a human passenger for the first time. T...

VLM-Mapper: Multimodal vision-language models for prior guided lane-level HD map construction

The automated construction of High-Definition (HD) maps from remote sensing data is essential for modern intelligent transportation systems and spatial data infrastructure. While imagery provides a scalable solution for lane-level HD map construction, traditional discriminative models often struggle in complex scenario...

Discovering session-level demand patterns for reinforcement learning-based dynamic inventory allocation in live-streaming e-commerce

Pre-control scheme generation for extreme-weather power imbalance risk using multi-resource frequency regulation and deep reinforcement learning

A pre-control scheme generation method based on the integration of multiple frequency regulation resources and deep reinforcement learning that achieves a total cost of 120.6 CNY, a success rate of 94.7%, and a violation rate of 0.0% under extreme weather scenarios are proposed.

Machine Learning for Logic Gate Synthesis and Optimization in Electronic Design Automation

The increasing complexity of integrated circuits has made logic synthesis and gate-level optimization important bottlenecks in electronic design automation (EDA). Conventional synthesis flows rely on deterministic transformations, handcrafted heuristics, and repeated evaluation of large design spaces. Machine learning...

Exploration Cost: The Exploration Coefficient in Onboard Reinforcement-Learning Schedulers Is an Independent Driver of Battery Aging

Reinforcement-learning (RL) schedulers are now flying on real spacecraft: NASA’s Carruthers Geocorona Observatory (launched 24 September 2025, per NASA’s own mission status page — not itself a LEO mission; it orbits near the Sun–Earth L1 point) uses deep RL as its default operational scheduler for long-horizon operatio...