Special Issue "Philosophy and Epistemology of Deep Learning"
A special issue of Philosophies (ISSN 2409-9287).
Deadline for manuscript submissions: 15 February 2019
Dr. Hector Zenil
1. Algorithmic Dynamics Lab, Unit of Computational Medicine, Science for Life Laboratory, Center of Molecular Medicine, Karolinska Institute, Stockholm, Sweden. 2. Algorithmic Nature Group, Laboratory of Scientific Research for the Natural and Digital Sciences, Paris, France. 3. Oxford Immune Algorithmics, Oxford University Innovation, Oxford, U.K.
Interests: algorithmic learning; information theory; complex systems; philosophy of information; randomness; systems biology; digital philosophy; reprogrammability; cellular automata; measures of sophistication
Prof. Dr. Selmer Bringsjord
Current popular approaches to Machine Learning (ML), Deep Learning (DL) and Artificial Intelligence (AI) are mostly statistic in nature and are not well equipped to deal with abstraction and explanation. In particular, they cannot generate candidate models or make generalizations directly from data to discover possible causal mechanisms. One way that researchers are using to discover how deep learning algorithms work is by using what are called ‘generative models’ (a possible misnomer). They train a learning algorithm and handicap it systematically whilst asking it to generate examples. By observing the resulting generated examples they can make inferences of what may be happening in the algorithm at some level. However, the current trends and methods are widely considered black-box approaches that have work amazingly well in tasks of classification but provide little to none understanding of causation and are unable to deal with symbolic computation such as logical inference and explanation. As a consequence, they also fail to be scalable in domains for which they were not trained for, and they require tons of data to be trained before doing anything interesting requiring training every time they are presented with (even slightly) different data. Moreover, an open question is how other cognitive features in humans such as consciousness may be related to current and future directions in deep learning and whether such features may provide advantages or disadvantages. The aim of this special issue is thus to attempt to ask the right questions and shed some light on the achievements, limitations and future directions in reinforcement/deep learning approaches and differentiable programming, in particular at the interplay of data and model-driven directions beyond the current state mostly based on traditional statistics, whether a fundamental theory is needed or already exists and the implications that current and future technology based on deep learning and differentiable programming may have in science, technology and society.
Dr. Hector Zenil
Prof. Dr. Selmer Bringsjord
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 papers will be 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. Philosophies is an international peer-reviewed open access quarterly journal published by MDPI.
Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) is waived for well-prepared manuscripts submitted to this issue. 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.