Keynotes

To register for free for the virtual AGI-20 conference, now extended to June 22-26, 2020, please submit your email on this Google form.

PLEASE REFER TO THE SCHEDULE PAGE FOR THE LATEST SCHEDULING UPDATES

Keynotes by:

  • Dr. Ben Goertzel, SingularityNET Foundation CEO
  • Dr. Valery Tarasov, Bauman Moscow State Technical University
    • Title: From Psychonics to Artificial General Intelligence and Hybrid Intelligence
  • Dr. Joscha Bach, AI Foundation
  • Dr. Anton Kolonin, SingularityNET Foundation
    • Title: On Global Brain and AGI co-evolution
  • Dr. David Hanson, Hanson Robotics
    • Title: Patternist Ethics, Empathic Robotics and Beneficial AGI
  • Dr. Andrey Chertok, Sberbank
    • Title: Digital Manager
  • Guest Lecture: Tatiana Shavrina, Sberbank
    • Title: Russian SuperGLUE
  • Guest Lecture: Ibby Benali, SingularityNET Foundation
    • Title: Increasing Global Collective Intelligence to Combat COVID-19

Abstract of Dr. Tarasov’s Keynote

The objective of the talk consists in considering Russian historical prerequisites of Artificial General Intelligence (AGI) related to Pospelov’s Psychonics, Venda’s Hybrid Intelligence (HI) and our Synergetic Artificial Intelligence (SynAI), as well as presenting modern AGI challenges by taking the example of Collaborative Robotics.  

So the talk contains two main parts. At first, some reasons for the rising interest to holistic AI models are discussed. Basic ideas and principles of Psychonics are presented as an early mentalist version of AGI. By analogy with Bionics Prof. Dmitry Pospelov has coined the term Psychonics to specify a sort of Psychology for Artificial Systems. The objective of Psychonics is the investigation of basic psychological approaches, methods, models and tools in order to provide their practical use in AI to create advanced intelligent systems. Here some results concerning the architecture of human psyche, organization of human activity and features of human personality are of primary concern. 

 The paradigm of Psychonics may be viewed as a direct precursor of AGI. Indeed, already in 1982 D.A.Pospelov formulated in his book «Fantasy or Science: on the Way to Artificial Intelligence» a very wide vision of Artificial Intelligence: «Modern investigations in AI should be aimed at studying and modeling of human psyche for the sake of its simulation in technical systems to solve a variety of practical problems conventionally viewed as intelligent ones» (p.211).

At the second part of the talk, the challenges of collaborative robotics associated with the need in constructing autonomous, safe, easy to control robots in order to establish a real partnership in human-robot systems are considered. An architecture of collaborative robot as a cognitive, communicative, understanding agent is introduced. The concepts of interactive dialogical control and multi-modal personalized anthropocentric interfaces to enable human-robot collaboration are presented. Apart visual, audio tactile components, such a friendly interface ought to have mimic and gesture components to enable some modes of easy robot’s visual control and learning by demonstration. The direct and inverse problems of Emotion Engineering in Collaborative Robotics together with the formal condition of emotional consonance are stated. An example of facial emotion recognition with using convolutional neural network in the context of robot’s performance evaluation and understanding is given.