Multi-Task Multi-Agent Reinforcement Learning With Interaction and Task Representations.

Multi-task multi-agent reinforcement learning (MT-MARL) is capable of leveraging useful knowledge across multiple related tasks to improve performance on any single task. While recent studies have tentatively achieved this by learning independent policies on a shared representation space, we pinpoint that further advancements can be realized by explicitly characterizing agent interactions within these multi-agent tasks and identifying task relations for selective reuse. To this end, this article proposes Representing Interactions and Tasks (RIT), a novel MT-MARL algorithm that characterizes both intra-task agent interactions and inter-task task relations.

State Abstraction via Deep Supervised Hash Learning.

State abstraction is a widely used technique in reinforcement learning (RL) that compresses the state space to accelerate learning algorithms. However, designing an effective abstraction function in large-scale or high-dimensional state space problems remains a significant challenge. In this brief, we present a novel state abstraction method based on deep supervised hash learning (DSH) and provide a theoretical analysis of its near-optimal property.

Enhancement of the Electric-Force Response of Carbon Black/Silicone Rubber Composites by Silane Coupling Agents.

Flexible strain sensors have a wide range of applications in the field of health monitoring of seismic isolation bearings. However, the nonmonotonic response with shoulder peaks limits their application in practical engineering. Here we eliminate the shoulder peak phenomenon during the resistive-strain response by adjusting the dispersion of conductive nanofillers.

Tracking an Fe (O) Intermediate for Water Oxidation in Water.

High-valent iron-oxo species are appealing for conducting O-O bond formation for water oxidation reactions. However, their high reactivity poses a great challenge to the dissection of their chemical transformations. Herein, we introduce an electron-rich and oxidation-resistant ligand, 2-[(2,2'-bipyridin)-6-yl]propan-2-ol to stabilize such fleeting intermediates.

Direct Excitation of Aldehyde to Activate the C(sp )-H Bond by Cobaloxime Catalysis toward Fluorenones Synthesis with Hydrogen Evolution.

A new way to form fluorenones via the direct excitation of substrates instead of photocatalyst to activate the C(sp )-H bond under redox-neutral condition is reported. Our design relies on the photoexcited aromatic aldehyde intermediates that can be intercepted by cobaloxime catalyst through single electron transfer for following β-H elimination. The generation of acyl radical and successful interception by a metal catalyst cobaloxime avoid the use of a photocatalyst and stoichiometric external oxidants, affording a series of highly substituted fluorenones, including six-membered ketones, such as xanthone and thioxanthone derivatives in good to excellent yields, and with hydrogen as the only byproduct.