AGI-10 Press Release
Our Invited Speakers page
Presenters of papers will have 15 minutes; presenters of position statements 3 minutes. Sessions will also finish with a group discussion of about 25 minutes.
Posters should ideally be on A0 paper.
Friday, March 5
PRE-CONFERENCE TUTORIALS, 10:30 – 17:20
- Marcus Hutter (ANU): Universal Artificial Intelligence, 10:30-11:30
- Moshe Looks (Google): Automated Program Learning, 11:45-12:45
- Ben Goertzel (Novamente LLC & Xiamen University): Using Virtual Agents and Physical Robots for AGI Research, 14:45-15:45
- Randal Koene (Tecnalia): Neural Mechanisms of Reinforcement Learning: How Spreading Activity in the Prefrontal Cortex can implement Reinforcement Learning for Goal-Directed Tasks, 16:00-17:00.
OPENING RECEPTION, 18:00 – 19:30 (outside the auditorium)
(A large buffet meal to welcome the conference)
KEYNOTE ADDRESS, 19:45 – 21:15
Richard Sutton, University of Alberta
Richard Sutton, author with Andrew Barto of Reinforcement Learning: an Introduction, is a Professor at the University of Alberta and one of the key figures in the field of artificial intelligence and learning.
Saturday, March 6
Intro: Past & Future of AGI, 8:30 – 9:30
- Ben Goertzel, Marcus Hutter, Juergen Schmidhuber
Session 1: AGI Evaluation, Chair: Moshe Looks, 9:30 – 11:00
- The Evaluation of AGI Systems
Pei Wang - What we Might Look for in an AGI Benchmark
Brandon Rohrer - On Evaluating Agent Performance in a Fixed Period of Time
Jose Hernandez-Orallo - The Toy Box Problem (and a Preliminary Solution)
Benjamin Johnston
Short position statements:
- A (hopefully) Unbiased Universal Environment Class for Measuring Intelligence of Biological and Artificial Systems
Jose Hernandez-Orallo - A Theoretical Framework to Formalize AGI-Hard Problems
Pedro Demasi, Jayme Szwarcfiter, Adriano Cruz
INVITED LECTURE, 11:30 – 13:00
Randal Koene, Whole Brain Emulation: Issues of scope and resolution, and the need for new methods of in-vivo recording
Randal Koene is the director of the Department of Neuroengineering, Fatronik-Tecnalia Foundation. His research objective, Whole Brain Emulation, is very exciting to the AGI field.
Session 2: Cognitive Architectures, Chair: Benjamin Johnston, 14:15 – 15:45
- Cognitive Architecture Requirements for Achieving AGI
John Laird, Robert Wray - The CHREST Architecture of Cognition: The Role of Perception in General Intelligence
Fernand Gobet, Peter Lane - A Cognitive Architecture for Knowledge Exploitation
Gee Wah Ng, Yuan Sin Tan, Loo Nin Teow, Khin Hua Ng, Kheng Hwee Tan, Rui Zhong Chan - Sketch of an AGI Architecture with Illustration
Andras Lorincz, Zoltan Bardosi, Daniel Takacs
Short position statement:
- Concept Formation in the Ouroboros Model
Knud Thomsen
Session 3: Adaptive Agents, Chair: Kai-Uwe Kühnberger, 16:15 – 18:15
- A Bayesian Rule for Adaptive Control based on Causal Interventions
Pedro Ortega, Daniel Braun - A Generic Adaptive Agent Architecture Integrating Cognitive and Affective States and their Interaction
Zulfiqar A. Memon, Jan Treur - Grounding Possible Worlds Semantics in Experiential Semantics
Matthew Ikle, Ben Goertzel - A General Intelligence Oriented Architecture for Embodied Natural Language Processing
Ben Goertzel, Cassio Pennachin, Samir Araujo, Ruiting Lian, Fabricio Silva, Murilo Queiroz, Welter Silva, Mike Ross, Linas Vepstas, Andre Senna - Toward a Formal Characterization of Real-World General Intelligence
Ben Goertzel
Short position statements:
- Relational Local Iterative Compression
Laurent Orseau - A Minimum Relative Entropy Principle for AGI
Antoine van de Ven, Ben Schouten - Software Design of an AGI System Based on Perception Loop
Antonio Chella, Massimo Cossentino, Valeria Seidita
Poster Session, 20:00 – 23:00
Poster presentation of short position statements
Sunday, March 7
Session 4: Universal Search, Chair: Shane Legg, 8:30 – 10:00
- Towards Practical Universal Search
Tom Schaul, Juergen Schmidhuber - Frontier Search
Yi Sun, Tobias Glasmachers, Tom Schaul, Juergen Schmidhuber - Searching for Minimal Neural Networks in Fourier Space
Jan Koutnik, Faustino Gomez, Juergen Schmidhuber - Algorithmic Probability, Heuristic Programming and AGI
Ray Solomonoff (read by Marcus Hutter)
Short position statement:
- Stochastic Grammar Based Incremental Machine Learning Using Scheme
Eray Ozkural, Cevdet Aykanat
Session 5: Aspects of Reasoning, Chair: Kristinn Thorisson, 10:30 – 12:30
- Discovering and Characterizing Hidden Variables
Soumi Ray, Tim Oates - Remarks on the Meaning of Analogical Relations
Ulf Krumnack, Helmar Gust, Angela Schwering, Kai-Uwe Kuehnberger - Quantitative Spatial Reasoning for General Intelligence
Unmesh Kurup, Nicholas Cassimatis - Efficient Constraint-Satisfaction in Domains with Time
Perrin Bignoli, Nicholas Cassimatis, Arthi Murugesan - Playing General Structure Rewriting Games
Lukasz Kaiser, Lukasz Stafiniak - A Conversion Between Utility and Information
Pedro Ortega, Daniel Braun
Short position statement:
- Uncertain Spatiotemporal Logic for General Intelligence
Nil Geisweiller, Ben Goertzel
Session 6: Machine Learning, Chair: Josh Hall, 14:00 – 15:30
- An Artificial Intelligence Model that Combines Spatial and Temporal Perception
Jianglong Nan, Fintan Costello - Artificial General Segmentation
Daniel Hewlett, Paul Cohen - GQ(λ): A General Gradient Algorithm for Temporal-Difference Prediction Learning with Eligibility Traces
Hamid Maei, Richard Sutton - Towards Automated Code Generation for Autonomous Mobile Robots
Dermot Kerr, Ulrich Nehmzow, Stephen A. Billings
Session 7: Visions & Ideas, Chair: Ben Goertzel, 16:00 – 17:10
- Artificial Scientists & Artists Based on the Formal Theory of Creativity
Juergen Schmidhuber - Designing a Safe Motivational System for Intelligent Machines
Mark Waser
Short position statements:
- Compression-Driven Progress in Science
Leo Pape - Compression Progress, Pseudorandomness, & Hyperbolic Discounting
Moshe Looks - Neuroethological Approach to Understanding Intelligence
DaeEun Kim - On Super-Turing Computing Power and Hierarchies of Artificial General Intelligence Systems
Jiri Wiedermann
AGI System Demonstrations, 17:30 – 19:00
“AGI Machines”
Monday, March 8
WORKSHOP, 8:30 – 11:40
“Toward a Serious Computational Science of Intelligence”
- SCSI Overview & Introduction, 8:30 – 8:45
Selmer Bringsjord & Naveen Sundar G. - In Memoriam: John Pollock, AGI, and SCSI, 8:45 – 9:00
Selmer Bringsjord - My Approach to, and Thoughts on, SCSI — Encapsulated, 9:00 – 9:30
Marcus Hutter - How should we get serious about developing AGI? 9:30 – 10:00
Shane Legg - Theories of Artificial Intelligence — Meta-theoretical considerations, 10:30 – 11:00
Pei Wang - Open Discussion & Debate, 11:00 – 12:00
moderated by Selmer Bringsjord
Coffee break, 10:00-10:30
BUSINESS MEETING, 12:00-13:00
- Next: Journal issues, etc.
- AGI-11, AGI-12 and beyond
- Journal of AGI
- AGI-Network
- Other issues relevant to the AGI community raised by participants
Lunch in the university cafeteria (with voucher), 13:15-14:00
FINAL WORKSHOP, 14:00- 17:00 (open ended)
“Roadmaps to AGI and the Future of AGI”
- An Overview of Models of Technological Singularity,
Anders Sandberg - Symbiotic AGI and Emergent Intelligence: Is a Planet Enough?
David Orban - Bootstrapping safe AGI goal systems
Roko Mijic - Remarks on AGI Ethics
Mark Waser - A Roadmap to Human-Level AGI: Report of the Conclusions of the Oct 2008 “AGI Roadmap Workshop”,
Ben Goertzel

Pictures from the University of Lugano conference venue © Università della Svizzera italiana.