Bayesian Inference: Algorithms and Challenges in Machine Learning

Dear Colleagues,

We are inviting submissions to a Special Issue on “Bayesian Inference: Algorithms and Challenges in Machine Learning”. Algorithms based on the Bayesian interpretation of probability have lent themselves well to several machine learning paradigms. Bayesian inference is mainly based on a distribution over model parameters which indicates levels of credibility for any setting given the observed data. Bayesian methods are also rigorous in quantifying (predictive) uncertainty, which is an issue of paramount importance in the quest to adopt automated decision-making systems in applications like self-driving cars and medical diagnosis. Hence, Bayesian methods and inference represent a general framework that can be leveraged and built upon in many machine learning domains. In this Special Issue, we invite submissions to algorithms and models which base their reasoning on Bayesian inference. We welcome original research, in the form of methodological and/or theoretical studies, as well as state-of-the-art review papers. Application-oriented studies can also fit if the Bayesian treatment of the application depicts sufficient novelty.

Topics of interest include (but are not limited to) the following:

• Approximate inference algorithms: Variational inference, Markov chain Monte Carlo (MCMC)-based inference, etc.
• Bayesian neural networks: Modelling epistemic and/or aleatoric uncertainty, scalability issues, etc.
• Bayesian approaches to fairness, interpretability, and robustness in machine learning.
• Bayesian approaches to transfer learning and related paradigms, e.g., domain adaptation, meta-learning, lifelong learning and multi-task learning.
• Probabilistic graphical models (PGMs): Bayesian networks, undirected graphical models, etc.
• Bayesian optimization: Gaussian processes (GPs), acquisition functions, etc.
• Likelihood-free inference.

We look forward to receiving your contributions.

Dr. Tameem Adel
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. Mathematics is an international peer-reviewed open access semimonthly 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 2600 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.

• variational Bayesian methods
• Markov chain Monte Carlo (MCMC) methods
• Bayesian neural networks
• Bayesian optimization
• Bayesian approaches for meta-learning and lifelong learning
• Bayesian reinforcement learning
• Bayesian networks
• probabilistic graphical models
• approximate Bayesian computation (ABC) algorithms
• Bayesian approaches for developing interpretable machine learning models