From Cognitive Bias to Whole Brain Emulation. The Perspective of Algorithmic Information -- A Shared Perspective -- The Ten Seminars -- On Common Ground -- The Present and the Past -- The First Discussion: On Communication -- The Historical Seminar: The Deeply Engrained Worship of Tidy-Looking Dichotomies -- Seminar 10 Algorithmic Growth -- 1.1. Information? What Information? -- The Second Discussion: On Complexity -- Seminar 2 From Algorithmic Growth to Endpoint Information -- 1.2. Noise and Relevant Information -- The Third Discussion: On Apple Trees and the Immune System -- Seminar 3 From Randomness to Precision -- 1.3. Autonomous Agents and Local Rules -- The Fourth Discussion: On Filopodia and Soccer Games -- Seminar 4 From Local Rules to Robustness -- 2. Of Players And Rules -- 2.1. The Benzer Paradox -- The Fifth Discussion: On the Genetic Encoding of Behavior -- Seminar 5 From Molecular Mechanisms to Evolutionary Programming -- 2.2. The Molecules That Could -- The Sixth Discussion: On Guidance Cues and Target Recognition -- Seminar 6 From Chemoaffinity to the Virtues of Permissiveness -- 2.3. The Levels Problem -- The Seventh Discussion: On Context -- Seminar 7 From Genes to Cells to Circuits -- 3. Brain Development And Artificial Intelligence -- 3.1. You Are Your History -- The Eighth Discussion: On Development and the Long Reach of the Past -- Seminar 8 From Development to Function -- 3.2. Self-Assembly versus "Build First, Train Later" -- The Ninth Discussion: On the Growth of Artificial Neural Networks -- Seminar 9 From Algorithmic Growth to Artificial Intelligence -- 3.3. Final Frontiers: Beloved Beliefs and the AI-Brain Interface -- The Tenth Discussion: On Connecting the Brain and AI -- Seminar 10 From Cognitive Bias to Whole Brain Emulation.
Summary:
"In this book, Peter Robin Hiesinger explores historical and contemporary attempts to understand the information needed to make biological and artificial neural networks. Developmental neurobiologists and computer scientists with an interest in artificial intelligence - driven by the promise and resources of biomedical research on the one hand, and by the promise and advances of computer technology on the other - are trying to understand the fundamental principles that guide the generation of an intelligent system. Yet, though researchers in these disciplines share a common interest, their perspectives and approaches are often quite different. The book makes the case that "the information problem" underlies both fields, driving the questions that are driving forward the frontiers, and aims to encourage cross-disciplinary communication and understanding, to help both fields make progress. The questions that challenge researchers in these fields include the following. How does genetic information unfold during the years-long process of human brain development, and can this be a short-cut to create human-level artificial intelligence? Is the biological brain just messy hardware that can be improved upon by running learning algorithms in computers? Can artificial intelligence bypass evolutionary programming of "grown" networks? These questions are tightly linked, and answering them requires an understanding of how information unfolds algorithmically to generate functional neural networks. Via a series of closely linked "discussions" (fictional dialogues between researchers in different disciplines) and pedagogical "seminars," the author explores the different challenges facing researchers working on neural networks, their different perspectives and approaches, as well as the common ground and understanding to be found amongst those sharing an interest in the development of biological brains and artificial intelligent systems"-- Provided by publisher.
This resource is supported by the Institute of Museum and Library Services under the provisions of the Library Services and Technology Act as administered by State Library of Iowa.