Обзор конференций Artificial general intelligence, AGI Safety & Impacts AGI’ 12 @ Oxford А. С. Потапов
Artificial General Intelligence AGI Sessions • Cognitive Architectures & Models A, B, C • Universal Intelligence and its Formal Approximations • Mathematical Formalisms and Tools • Conceptual and Contextual Issues • Special Session on AGI & Neuroscience 2
Artificial General Intelligence Keynotes • Margaret Boden (University of Sussex) Creativity and AGI • Angelo Cangelosi (University of Plymouth) From Sensorimotor Intelligence to Symbols: Developmental Robotics Experiments • David Hanson (Hanson Robotics) Open source genius machines who care: Growing Friendly AGI via GENI, Glue and Lovable Characters • Nick Bostrom (Future of Humanity Institute) The Superintelligence Control Problem 3
From Sensorimotor Intelligence to Symbols: Developmental Robotics Experiments 4
5
6
7
Open source genius machines who care: Growing Friendly AGI via GENI, Glue and Lovable Characters 8
9
Когнитивные архитектуры Доклады • Joscha Bach. The Next Generation of the Micro. Psi Framework • Pei Wang. Motivation Management in AGI Systems • Helgi Helgason, Eric Nivel and Thórisson Kristinn. On Attention Mechanisms for AGI Architectures: A Design Proposal • Alessandro Oltramari and Christian Lebiere. Pursuing Artificial General Intelligence By Leveraging the Knowledge Capabilities Of ACT-R • Peter Lane and Fernand Gobet. CHREST models of implicit learning and board game interpretation 10
Когнитивные архитектуры: Micro. Psi 11
12
13
Когнитивные архитектуры Доклады • Ben Goertzel. Perception Processing for General Intelligence: Bridging the Symbolic/Subsymbolic Gap • Jade O’Neill and Ben Goertzel. Pattern Mining for General Intelligence: The FISHGRAM Algorithm for Frequent and Interesting Subhypergraph Mining • Ruiting Lian and Ben Goertzel. Syntax-Semantic Mapping for General Intelligence: Language Comprehension as Hypergraph Homomorphism, Language Generation as Constraint Satisfaction • Paul Rosenbloom. Deconstructing Reinforcement Learning in Sigma • Paul Rosenbloom. Extending Mental Imagery in Sigma 14
Когнитивные архитектуры: Open. Cog 15
16
17
Универсальный интеллект Доклады • Joel Veness, Peter Sunehag and Marcus Hutter. On Ensemble Techniques for AIXI Approximation • Peter Sunehag and Marcus Hutter. OPTIMISTIC AIXI • Laurent Orseau, Mark Ring. Memory issues of intelligent agents • Laurent Orseau, Mark Ring. Space-Time Embedded Intelligence • Alexey Potapov, Andrew Svitenkov and Yurii Vinogradov. Differences between Kolmogorov Complexity and Solomonoff Probability: Consequences for AGI • Alexey Potapov and Sergey Rodionov. Extending Universal Intelligence Models with a Formal Notion of Representation 18
19
20
21
22
23
24
Capability control • Boxing methods – Physical containment – Information containment • Incentive methods – Social integration – Reward button • Stunting • Tripwires 25
26
27
28
29
30
31
32
33
34
35
Whole Brain Emulation 36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
Выводы? 51
Спасибо за внимание 52
Скачать презентацию Обзор конференций Artificial general intelligence AGI Safety
seminarAI_34_AGI.ppt