Deep Blue system overview

Hsu, Feng-Hsiung; Campbell, Murray; Hoane, A. Joseph

doi:10.1145/224538.224567

Cited by 30 publications

(15 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We now know that these tasks are absolutely trivial for a machine and thus do not appear to test the machine's intelligence in any meaningful sense. Indeed even the mentally demanding task of playing chess can be largely reduced to brute force search [HCH95]. What else may in time be possible with relatively simple algorithms running on powerful machines is hard to say.…”

Section: Introductionmentioning

confidence: 99%

Universal Intelligence: A Definition of Machine Intelligence

2007

View full text Add to dashboard Cite

A fundamental problem in artificial intelligence is that nobody really knows what intelligence is. The problem is especially acute when we need to consider artificial systems which are significantly different to humans. In this paper we approach this problem in the following way: We take a number of well known informal definitions of human intelligence that have been given by experts, and extract their essential features. These are then mathematically formalised to produce a general measure of intelligence for arbitrary machines. We believe that this equation formally captures the concept of machine intelligence in the broadest reasonable sense. We then show how this formal definition is related to the theory of universal optimal learning agents. Finally, we survey the many other tests and definitions of intelligence that have been proposed for machines.

show abstract

Section: Introductionmentioning

confidence: 99%

Universal Intelligence: A Definition of Machine Intelligence

2007

View full text Add to dashboard Cite

show abstract

“…To compound these problems, most performance metrics have been measured with the myopic lookahead search of depth one. This is in contrast to game-playing practice where most competitive systems gain a substantial benefit from a deeper search horizon (Schaeffer, Culberson, Treloar, Knight, Lu, & Szafron, 1992;Hsu, Campbell, & Hoane, 1995;Buro, 1995).We take a step towards a unified view of learning in real-time search and make four contributions. First, we introduce a simple three-parameter framework (named LRTS) that includes LRTA*, -LRTA* , SLA* and γ-Trap as its special cases.…”

mentioning

confidence: 99%

Learning in Real-Time Search: A Unifying Framework

Bulitko¹,

Lee²

2006

jair

View full text Add to dashboard Cite

Real-time search methods are suited for tasks in which the agent is interacting with an initially unknown environment in real time. In such simultaneous planning and learning problems, the agent has to select its actions in a limited amount of time, while sensing only a local part of the environment centered at the agent's current location. Real-time heuristic search agents select actions using a limited lookahead search and evaluating the frontier states with a heuristic function. Over repeated experiences, they refine heuristic values of states to avoid infinite loops and to converge to better solutions. The wide spread of such settings in autonomous software and hardware agents has led to an explosion of real-time search algorithms over the last two decades. Not only is a potential user confronted with a hodgepodge of algorithms, but he also faces the choice of control parameters they use. In this paper we address both problems. The first contribution is an introduction of a simple three-parameter framework (named LRTS) which extracts the core ideas behind many existing algorithms. We then prove that LRTA*, -LRTA* , SLA*, and γ-Trap algorithms are special cases of our framework. Thus, they are unified and extended with additional features. Second, we prove completeness and convergence of any algorithm covered by the LRTS framework. Third, we prove several upper-bounds relating the control parameters and solution quality. Finally, we analyze the influence of the three control parameters empirically in the realistic scalable domains of real-time navigation on initially unknown maps from a commercial role-playing game as well as routing in ad hoc sensor networks. MotivationIn this paper, we consider a simultaneous planning and learning problem. One motivating application lies with navigation on an initially unknown map under real-time constraints. As an example, consider a robot driving to work every morning. Imagine the robot to be a newcomer to the town. The first route the robot finds may not be optimal because traffic jams, road conditions, and other factors are initially unknown. With the passage of time, the robot continues to learn and eventually converges to a nearly optimal commute. Note that planning and learning happen while the robot is driving and therefore are subject to time constraints.Present-day mobile robots are often plagued by localization problems and power limitations, but their simulation counter-parts already allow researchers to focus on the planning and learning problem. For instance, the RoboCup Rescue simulation league (Kitano, Tadokoro, Noda, Matsubara, Takahashi, Shinjou, & Shimada, 1999) requires real-time planning and learning with multiple agents mapping out an unknown terrain. Pathfinding is done in real time as various crises, involving fire spread and human victims trapped in rubble, progress while the agents plan.Similarly, many current-generation real-time strategy games employ a priori known maps. Full knowledge of the maps enables complete search methods such as A...

show abstract

“…Currently the two primary focuses of AI research are conventional (or symbolic) and computational; since intelligence still has such a broad definition, the two handles separate humanbased intelligence parts. Conventional AI, which entails rational logical reasoning based on a system of symbols representing human knowledge in a declarative form [50], has been used for such applications as chess games (reasoning) [51], conversation programs (text mining), 15 and for organizing domain-specific knowledge (expert systems) [52]. While conventional AI is capable of limited reasoning, planning, and abstract thinking, researchers acknowledge that the use of symbols does not represent "mindful" comprehension, and is limited in terms of learning from experience [53].…”

Section: Human-based Intelligencementioning

confidence: 99%

Toward the Human–Robot Co-Existence Society: On Safety Intelligence for Next Generation Robots

Weng¹,

Chen²,

Sun

2009

Int J of Soc Robotics

View full text Add to dashboard Cite

Technocrats from many developed countries, especially Japan and South Korea, are preparing for the human-robot co-existence society that they believe will emerge by 2030. Regulators are assuming that within the next two decades, robots will be capable of adapting to complex, unstructured environments and interacting with humans to assist with the performance of daily life tasks. Unlike heavily regulated industrial robots that toil in isolated settings, Next Generation Robots will have relative autonomy, which raises a number of safety issues that are the focus of this article. Our purpose is to describe a framework for a legal system focused on Next Generation Robots safety issues, including a Safety Intelligence concept that addresses robot Open-Texture Risk. We express doubt that a model based on Isaac Asimov's Three Laws of Robotics can ever be a suitable foundation for creating an artificial moral agency ensuring robot safety. Finally, we make predictions about the most significant Next Generation Robots safety issues that will arise as the human-robot co-existence society emerges.Y.-H. Weng ( ) Conscription Agency, Ministry of the Interior, Republic of China,

show abstract

Deep Blue system overview

Abstract: This paper gives an overview of the system, and examines the prospects of reaching the goal.

Cited by 30 publications

References 8 publications

Universal Intelligence: A Definition of Machine Intelligence

Universal Intelligence: A Definition of Machine Intelligence

Learning in Real-Time Search: A Unifying Framework

Toward the Human–Robot Co-Existence Society: On Safety Intelligence for Next Generation Robots

Contact Info

Product

Resources

About