I had to oppurtunity to deliver this keynote talk at ICAR-MEET 2020, the annual workshop of the Institute for High Performance Computing of the National Research Council of Italy (CNR).
Below the link to the slides:
Commonsense reasoning as a key feature for dynamic knowledge invention and computational creativity from Antonio Lieto
I am really glad to announce my forthcoming book "Cognitive Design for Artificial Minds" that will be published in 2021 by Taylor & Francis (T&F).
The book can already be pre-ordered in the main online bookshops (e.g. from Routledge (T&F) to Amazon Walmart Barnes& Noble etc.)
I am posting below the "blurb" and the Table of Contents of the Book from the Routledge website (T&F).
Cognitive Design for Artificial Minds explains the crucial role that human cognition research plays in the design and realization of artificial intelligence systems, illustrating the steps necessary for the design of artificial models of cognition. It bridges the gap between the theoretical, experimental and technological issues addressed in the context of AI of cognitive inspiration and computational cognitive science.
Beginning with an overview of the historical, methodological and technical issues in the field of Cognitively-Inspired Artificial Intelligence, Lieto illustrates how the cognitive design approach has an important role to play in the development of intelligent AI technologies and plausible computational models of cognition. Introducing a unique perspective that draws upon Cybernetics and early AI principles, Lieto emphasizes the need for an equivalence between cognitive processes and implemented AI procedures, in order to realise biologically and cognitively inspired artificial minds. He also introduces the Minimal Cognitive Grid, a pragmatic method to rank the different degrees of biologically and cognitive accuracy of artificial systems in order project and predict their explanatory power with respect to the natural systems taken as source of inspiration.
Providing a comprehensive overview of cognitive design principles in constructing artificial minds, this text will be essential reading for students and researchers of artificial intelligence and cognitive science.
Introduction
Chapter 1. Cognitive Science and Artificial Intelligence: Death and Rebirth of a Collaboration
1.1. When Cognitive Science was AI
1.2. From The General Problem Solver to the Society of Mind: cognitivist insights from the early AI era
1.3. Heuristics and AI Eras
1.4. Modelling Paradigms and AI Eras: Cognitivist and Emergentist Perspective
1.5. Death and Rebirth of a Collaboration
Chapter 2. Cognitive and Machine Oriented Approaches to Intelligence in Artificial Systems
2.1. Nature vs Machine Inspired Approaches to Artificial Systems
2.2. Functionalist vs Structuralist Design Approaches
2.3. Levels of Analysis of Computational Systems
2.4. The Space of Cognitive Systems
2.5. Functional and Structural Neural Systems
2.6. Functional and Structural Symbolic Systems
Chapter 3. Principles of the Cognitive Design Approach
3.1. Classical, Bounded and Bounded-Rational Models of Cognition
3.2. Resource-Rationality Models
3.3 Kinds of Explanations
3.4 Levels of Plausibility and the Minimal Cognitive Grid (MCD)
Chapter 4. Examples of Cognitively Inspired Systems and application of the Minimal Cognitive Grid
4.1 Modern AI Systems: Cognitive Computing?
4.2 Cognitive Architectures
4.3 SOAR
4.4. ACT-R
4.5 Two Problems for the Knowledge Level in Cognitive Architectures
4.6. Knowledge Size and Knowledge Heterogeneity in SOAR and ACT-R
4.7. DUAL PECCS
Chapter 5. Evaluating the Performances of Artificial Systems
5.1. "Thinking" Machines and Turing Test(s)
5.2. The Chinese Room
5.3. The Newell Tests for a Theory of Cognition
5.4. The Winograd Schema Challenge
5.5. DARPA Challenges, Robocup and Robocup@Home
5.6. Comparison
Chapter 6. The Next Steps
6.1. The Road Travelled
6.2. The Way Forward
6.3. Towards a Standard Model of Mind/Common Model of Cognition
6.4. Community
The post is to share an update about an (unexpected) outcome of one of my main research outputs: the Heterogeneous Proxytypes (HP) hypothesis.
A brief recap before going to the point: back in 2014 I proposed heterogeneous proxytypes as a framework (alternative and integrative with respect to other approaches based on prototypes, exemplars, model-theory etc.) for conceptual representation and processing in cognitive science and artificial intelligence research (here the original paper http://www.sciencedirect.com/science/article/pii/S1877050914015233, presented at BICA 2014 held at the MIT in Boston).
Such a proposal has led to the design and development of categorization systems like DUAL-PECCS (the main references of such a cognitively-inspired knowledge representation and processing framework, and of its computational counterpart, are reported below):
- Antonio Lieto, "A Computational Framework for Concept Representation in Cognitive Systems and Architectures: Concepts as Heterogeneous Proxytypes" in Proceedings of BICA 2014, 5th Int. Conference of Biologically Inspired Cognitive Architectures, Boston, Massachusetts Institute of Technology (MIT), USA, 7-9 November 2014. Elsevier Procedia Computer Science, Vol. 41 (2014), pp. 6-14.
- Antonio Lieto, Daniele P. Radicioni and Valentina Rho, A Common-Sense Conceptual Categorization System Integrating Heterogeneous Proxytypes and the Dual Process of Reasoning". In Proceedings of the International Joint Conference on Artificial Intelligence (IJCAI), Buenos Aires, July 2015, pp. 875-881. AAAI press.
- Antonio Lieto, Daniele P. Radicioni and Valentina Rho "Dual PECCS: A Cognitive System for Conceptual Representation and Categorization" in Journal of Experimental & Theoretical Artificial Intelligence (JETAI), Vol. 29(2), pp. 433-452.
- Antonio Lieto, "Heterogeneous Proxytypes Extended: Integrating Theory-like Representations and Mechanisms with Prototypes and Exemplars", in Proceedings of BICA 2018, Springer Advances in Intelligent Systems and Computing, 2019.
The fact: in the past few days, thanks to a late notification by Google Scholar, I have discovered that HP hypothesis has been proposed, in a 2017 Ph.D Thesis at the Faculty of Education at the University of Johannesburg entitled "Towards a Science Education Learning Environment for Student Teachers of the Foundation Phase" (by E.C.A. Kok), as a tool for "foundation phase school teachers" (that I have just discovered being "school teachers teaching children the foundations of reading, writing and literacy and are also responsible for helping children to develop their thinking skills") to reflect on what kind of stimulus to present in order to let the students to activate the "right proxy-representation" in their working memory (by using the HPH terminology) as a trigger for letting learning take place.
Below an excerpt directly taken from the above-mentioned Ph.D. Thesis
 |
| First diagram sketch of the heterogeneous proxytypes hypothesis (on a croissant take-away "envelope"). Click to see the entire picture. |
Alphabet sembra aver associato un po' di conosenza al "mio" grafo (che visualizzo quando sono loggato nell'account Google).
I was invited to give a talk on the cognitive paradigm in AI research by the National University of "Kyiv-Mohyla Academy" in Kiev. Due to the COVID-19 crisis, the lecture was given online. The video of the online lecture is available below.
A shortlist of relevant insights from the book Ranking: The Unwritten Rules of the Social Game We All Play by Péter Érdi
