Reinforcement Learning: Algorithms and Applications

Dear Colleagues,

Reinforcement learning is a research area with tremendous progress in addressing the various fundamental challenges that arise in applications—yet wide-spread rollout requires overcoming additional translational challenges to put algorithms into practice. The aim of this Special Issue is to (1) identify and characterise rising fundamental and translational challenges, (2) delineate progress towards solving them, and (3) highlight insights gained in successful real-world applications. 

We invite high quality original contributions to this Special Issue on "Reinforcement Learning: Algorithms and Applications." While reinforcement learning has made large strides when simulations are available, this Special Issue particularly seeks to advance a holistic system view, including the application task modeling or model-learning stages (e.g., discussing bias-variance tradeoffs) or other methods that close the gap between theory and application. Submissions that integrate progress against a fundamental challenge with empirical evidence in a specific application are particularly encouraged. 

Reinforcement learning algorithms and applications may address the following (non-exhaustive) list of topics of interests:

• Fundamental and translational challenges, such as autonomy and interdependence, generalisation and bias, sample efficiency vs data availability, partial observability, the need for exploration, and scaling to large state/action/belief/parameter spaces.
• Reinforcement learning techniques, such as learning targets and backup operators, auxiliary tasks, hierarchical models, particular or multiple objectives (e.g., considering risk, robustness, safety, user preference elicitation, or explainability), multi-agent learning/population-based training, or multi-stage training procedures that generalise to the real world.
• Applications of functioning learning system and synergies of techniques to compose them, e.g., for control (robotics, smart industry), resource allocation (markets, negotiation, finance, energy, advertisements, transportation), or information systems (recommender systems, communication).
• Insights on performance, convergence, optimality, regret, topologies of representation/encoding of states, beliefs, policies etc., and their impact on applications. 

Contributions on deep learning are not required but explicitly invited, and expected to be contrasted to previously existing techniques.

Dr. Michael Kaisers
Guest Editor

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Algorithms is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 1600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• Learning algorithms 
• Applications of reinforcement learning 
• Planning in reinforcement learning tasks 
• Learning systems 
• Multi-agent learning 
• Auxiliary tasks 
• Deep learning