Entropy-Based Uncertainty Management Methods in Deep Learning
A special issue of Entropy (ISSN 1099-4300). This special issue belongs to the section "Multidisciplinary Applications".
Deadline for manuscript submissions: closed (29 February 2024) | Viewed by 479
Special Issue Editors
Interests: analytics and information fusion, BigData, machine, and deep Learning
Special Issues, Collections and Topics in MDPI journals
Interests: uncertainty measure; Shannon entropy; Tsallis entropy; Renyi entropy; Deng entropy; evidence theory; fuzzy sets; fractal; complex network; time series
Special Issues, Collections and Topics in MDPI journals
Special Issue Information
Dear Colleagues,
Uncertainty widely exists when humans observe-orient-decide-act (OODA) in the real world. Understanding and managing uncertainty is essential in an effective OODA decision-making loop. Quantifying uncertainty has been the focus of many scholars who have put forward various theories besides Bayesian, such as evidence or belief theory and its variants, fuzzy set, possibility theory, rough set, D and Z number theories, logical approaches, and hybridization of these methods. The Internet-of-Things (IoT) and Big Data era have brought about a deluge of information that challenge effective decision-making because of information-attached imperfections, such as completeness, availability, reliability, vagueness, conflict, and false information.
Several machine learning approaches have been developed using artificial intelligence (AI), and recently amongst these approaches, deep learning has surfaced as a promising method to solve complex problems involving high-dimensional data processing. However, managing and quantifying different types of uncertainty in machine learning/deep learning has not received much attention as of yet. While neural network technologies dominate deep learning, entropy-based approaches dominate the uncertainty management domain. Therefore, the goal of this Special Issue is to discuss recent development in uncertainty management in the field of deep learning applications such as intelligent systems, automatic control, natural language processing, computer vision and speech understanding, analytics and data mining, smart planning, robotics, information fusion, etc.
Contributions might address a mixture of deep learning technologies, taking into account uncertainty representation and reasoning. Some of the potential topics include, but not limited to, the following:
- Evidential deep neural networks (DNN);
- Rough DNN;
- DNN and fuzzy logic;
- Possibilistic DNN;
- Neural network-based uncertainty quantification;
- Uncertainty in big data analytics;
- Uncertainty in information fusion;
- Deep reinforcement learning;
- Bayesian deep learning;
- Classification by deep learning;
- Deep belief networks;
- Fuzzy convolutional neural network;
- Uncertainty-aware deep classification;
- Rough sets and neural network;
- Graph neural networks ;
- Deep possibilistic clustering;
- Deep neural dynamic Bayesian networks.
Dr. Éloi Bossé
Prof. Dr. Yong Deng
Guest Editors
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. Entropy 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 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.
Keywords
- belief theory
- uncertainty reasoning
- uncertainty quantification
- decision making
- machine/deep learning
- evidence theory
- uncertainty measures
- Deng entropy
- Shannon entropy
- fuzzy entropy, possibilistic entropy
- Gini entropy
Benefits of Publishing in a Special Issue
- Ease of navigation: Grouping papers by topic helps scholars navigate broad scope journals more efficiently.
- Greater discoverability: Special Issues support the reach and impact of scientific research. Articles in Special Issues are more discoverable and cited more frequently.
- Expansion of research network: Special Issues facilitate connections among authors, fostering scientific collaborations.
- External promotion: Articles in Special Issues are often promoted through the journal's social media, increasing their visibility.
- e-Book format: Special Issues with more than 10 articles can be published as dedicated e-books, ensuring wide and rapid dissemination.
Further information on MDPI's Special Issue polices can be found here.