Bart Kosko

from Wikipedia, the free encyclopedia

Bart Andrew Kosko (born February 7, 1960 in Kansas City ) is an author and professor of electrical engineering and law at the University of Southern California (USC). Kosko is known for his research on fuzzy logic and neural networks , as well as the author of several non-fiction books on these and related topics of artificial intelligence .

Personal background

Bart Kosko holds a bachelor's degree in philosophy and economics from USC, a master's degree in applied mathematics from UC San Diego, a doctorate in electrical engineering from UC Irvine, and an online JD in law from online Concord Law School. He is a licensed attorney in California and with the US Federal Supreme Court and has served as the Legal Secretary for the Los Angeles District Attorney's Office. Kosko is a political and religious skeptic and co-editor of the libertarian magazine "Liberty" .

Works

Kosko's most popular book to date is the fuzzy-logical international bestseller . A new way of thinking in which he questions the entire foundation of Western thought built on Aristotelian logic and opposes it to fuzzy-logical thinking as an alternative. From his religion-skeptical point of view, belief in God is a neural effect, but he also criticizes the typically Western overemphasis on scientific thinking. According to this, through science , objectivity and logic we would promise ourselves a security that could not exist. In our culture a binary- logical thinking of “either-or” would dominate, with which we suggest that there are undoubted and final truths. According to Kosko, this leads to a “black-and-white thinking” with fading out of any shades of gray, which is far more unrealistic than it would allow statements about reality . From his perspective as a scientist, too, Asian thinking (e.g. Buddhism ), in which the world is not viewed in apparent opposites, is closer to reality.

Kosko also published short stories and the novel "NanoTime" about a possible third world war that will take place on two days in 2030. Kosko uses a minimalist prose style without commas.

Researches

Kosko's research focuses on the main areas of fuzzy logic, neural networks and noise . He introduced several new terms into fuzzy logic, such as a. the cognitive map , additive fuzzy systems, fuzzy-associative memories, variable neural-based fuzzy systems, the conditional variance of fuzzy systems, and the geometric view of finite fuzzy sets as points in hypercubes , as well as their significance in the controversy between fuzzy and Probability .

In the research area of ​​neural networks Kosko introduced techniques for unsupervised learning according to Hebb's learning rule , as well as the concept of bidirectional associative memory .

Web links

Individual evidence

  1. ^ "Liberty" magazine
  2. Kosko, Bart: fuzzy-logical. A New Way of Thinking , Carlsen 1993
  3. Fuzzy Cognitive Maps (PDF; 1.3 MB)
  4. Global Stability of Generalized Additive Fuzzy Systems, August 1998 (PDF; 479 kB)
  5. Unsupervised Learning in Noise, March 1990 (PDF; 1.4 MB)
  6. Bidirectional Associative Memories, February 1998 (PDF; 1.1 MB)