Search

This article discusses the immense opportunities as well as significant risks presented by the advancements in artificial intelligence, comparing its impact to that of the industrial revolution. The authors from the One Hundred Year Study on AI recommend ten priorities, including investing in scientific foundations, ensuring safety and equity, devising regulations, combating disinformation, and fostering academia-industry collaboration. Emphasizing both short- and long-term impacts, they highlight the importance of international diplomacy, monitoring economic effects, understanding societal implications, and prioritizing the prevention of catastrophic outcomes to harness AI's benefits while mitigating its risks. These recommendations aim to spark and guide discussions and actions among the AI community and society at large.

When mechanisms such as auctions, rating systems, and elections are run in a highly anonymous environment such as the Internet, a key concern is that a single agent can participate multiple times by using false identifiers. Such false-name manipulations have traditionally not been considered in the theory of mechanism design. In this article, we review recent efforts to extend the theory to address this. We first review results for the basic concept of false-name-proofness. Because some of these results are very negative, we also discuss alternative models that allow us to circumvent some of these negative results., Technologies such as the Internet allow many spatially distributed parties (or agents) to rapidly interact according to intricate protocols. Some of the most exciting applications of this involve making decisions [...]

AAAI was pleased to present the AAAI-08 Workshop Program, held Sunday and Monday, July 13-14, in Chicago, Illinois, USA. The program included the following 15 workshops: Advancements in POMDP Solvers; AI Education Workshop Colloquium; Coordination, Organizations, Institutions, and Norms in Agent Systems, Enhanced Messaging; Human Implications of Human-Robot Interaction; Intelligent Techniques for Web Personalization and Recommender Systems; Metareasoning: Thinking about Thinking; Multidisciplinary Workshop on Advances in Preference Handling; Search in Artificial Intelligence and Robotics; Spatial and Temporal Reasoning; Trading Agent Design and Analysis; Transfer Learning for Complex Tasks; What Went Wrong and Why: Lessons from AI Research and Applications; and Wikipedia and Artificial Intelligence: An Evolving Synergy., The Workshop on Advancements in POMDP Solvers brought together active researchers in the area of solving partially observable Markov decision processes (POMDPs). Participants discussed various approaches to solving POMDPs, and [...]

The article discusses bias found in AI algorithmic systems -- known as Algorithmic Bias -- which returned results with errors related to skin color or gender. One paper by researchers Joy Buolamwini and Timnit Gebru documented substantial discrepancies across skin types and the reliability with which three commercial face recognition systems could classify gender. A followup study by the same researchers found that those companies that were originally audited released new classifiers with reduced error rates while other companies had much higher error rates.

Sharing computational resources amortizes cost and improves utilization and efficiency. When agents pool their resources, each becomes entitled to a portion of the shared pool. Static allocations in each round can guarantee entitlements and are strategy-proof, but efficiency suffers because allocations do not reflect variations in agents' demands for resources across rounds. Dynamic allocation mechanisms assign resources to agents across multiple rounds while guaranteeing agents their entitlements. Designing dynamic mechanisms is challenging, however, when agents are strategic and can benefit by misreporting their demands for resources. In this paper, we show that dynamic allocation mechanisms based on max-min fail to guarantee entitlements, strategy-proofness or both. We propose the flexible lending (FL) mechanism and show that it satisfies strategy-proofness and guarantees at least half of the utility from static allocations while providing an asymptotic efficiency guarantee. Our simulations with real and synthetic data show that the performance of the flexible lending mechanism is comparable to that of state-of-the-art mechanisms, providing agents with at least 0.98x, and on average 15x, of their utility from static allocations. Finally, we propose the T -period mechanism and prove that it satisfies strategy-proofness and guarantees entitlements for T ≤ 2.

The article discusses the moral and ethical aspects of designing algorithms for allocating resources.

A set of players delegate playing a game to a set of representatives, one for each player. We imagine that each player trusts their respective representative’s strategic abilities. Thus, we might imagine that per default, the original players would simply instruct the representatives to play the original game as best as they can. In this paper, we ask: are there safe Pareto improvements on this default way of giving instructions? That is, we imagine that the original players can coordinate to tell their representatives to only consider some subset of the available strategies and to assign utilities to outcomes differently than the original players. Then can the original players do this in such a way that the payoff is guaranteed to be weakly higher than under the default instructions for all the original players? In particular, can they Pareto-improve without probabilistic assumptions about how the representatives play games? In this paper, we give some examples of safe Pareto improvements. We prove that the notion of safe Pareto improvements is closely related to a notion of outcome correspondence between games. We also show that under some specific assumptions about how the representatives play games, finding safe Pareto improvements is NP-complete.