The Blyth Institute, a think tank that explores the relationships between biology, cognitive science, and engineering, recently released the second issue of their new journal, Communications of the Blyth Institute. This issue focuses on the philosophy of science, with several papers touching on issues in mind and machine learning.

The lead paper for the issue is by Dr. Sam Rakover, Professor Emeritus at the University of Haifa in Israel, which discusses a new way of considering the connections between mental states, conscious mental states, and higher order models of consciousness.

Rather than taking a reductive approach and forcing all mental activities to relate explicitly to biological features and chemistry, Rakover takes a functional approach. He focuses on the phenomenon of consciousness itself and how it fits with our overall cognitive functioning. He presents an initial sketch of a methodology that allows a better conceptualization of the method by which mental states get moved in and out of consciousness.

This edition also includes a review of Scott D. G. Ventureyra ’s recent book On the Origin of Consciousness..Eschewing reductionist thinking, Ventureyra focuses on why consciousness arose instead of how. The review makes the point that efforts to cram all questions into the “how” category can cause us to miss the most important details of some phenomena. Asking the questions which are appropriate to the phenomena leads to real knowledge.

Other papers in this volume include:

• a paper describing how generalizations can be quantified
• a paper describing how problematic null hypotheses have led to faulty reasoning in several sciences
• a paper looking at complicated dilemmas in science education
• a paper looking at the very nature of truth itself
• a paper looking at improving the notation for partial derivatives
• a paper examining a simple dataset that is easy for humans to learn, but surprisingly difficult for computers, and
• a paper looking at what information theory tells us about the minimal requirements for self-replicating systems and their probabilities

See also: Machine learning dates back to at least 300 BC The key to machine learning is not machines but mathematics (Jonathan Bartlett)

and

Walter Bradley Center fellow Jonathan Bartlett discovers longstanding flaw in an aspect of elementary calculus