Michael van Lent

Learning procedural knowledge through observation

The research presented here describes a framework that provides the necessary infrastructure to learn procedural knowledge from observation traces annotated with goal transition information. One instance of a learning-by-observation system, called KnoMic (Knowledge Mimic), is developed within this framework and evaluated in a complex domain. This evaluation demonstrates that learning by observation can acquire procedural knowledge and can acquire that knowledge more efficiently than standard knowledge acquisition.

PERFECT RECALL AND PRUNING IN GAMES WITH IMPERFECT INFORMATION

Games with imperfect information are an interesting and important class of games. They include most card games (e.g., bridge and poker) as well as many economic and political models. Here we investigate algorithms for findi ng the simplest form of a solution (a pure‐strategy equilibrium point) to imperfect information games expressed in their extensive (game tree) form. We introduce to the artificial intelligence community a classic algorithm, due to Wilson, that solves one‐player games with perfect recall. Wilsons algorithm, which we call iMP‐minimax, runs in time linear in the size of the game‐tree searched. In contrast to Wilsons result, Koller and Meggido have shown that finding a pure‐strategy equilibrium point in one‐player games without perfect recall is NP‐hard. Here, we provide another contrast to Wilsons result–we show that in games with perfect recall but more than one player, finding a pure‐strategy equilibrium point, given that such an equilibrium point exists, is NP‐hard.

Perceptually driven cognitive mapping of urban environments

Perceptually Driven Cognitive Mapping Of Urban Environments Randall W . Hill, Jr. USC Institute for Creative Technologies 13274 Fiji Way, Suite 600 Marina del Rey, CA 90292-7008 1-310-574-7815 hill@ict.usc.edu Changhee Han USC Institute for Creative Technologies 13274 Fiji Way, Suite 600 Marina del Rey, CA 90292-7008 1-310-574-5700 changhee@ict.usc.edu Michael van Lent USC Institute for Creative Technologies 13274 Fiji Way, Suite 600 Marina del Rey, CA 90292-7008 1-310-574-5710 vanlent@ict.usc.edu

Reflective tutoring for immersive simulation

Reflection is critically important for time-constrained training simulations that do not permit extensive tutor-student interactions during an exercise. Here, we describe a reflective tutoring system for a virtual human simulation of negotiation. The tutor helps students review their exercise, elicits where and how they could have done better, and uses explainable artificial intelligence (XAI) to allow students the chance to ask questions about the virtual humans behavior.

Creating densely populated virtual environments

Few virtual environments are capable of supporting large numbers of autonomous agents (> 5000) with complex decision-making on a single machine. This demonstration depicts such an agent infrastructure set within a game-based virtual environment. The embodied agent framework consists of two primary components: a lower-level navigation layer consisting of commercially-available AI middleware, and a higher-level cellular automata system driven by agent goals, resources and thresholds. The overarching game-based infrastructure consists of these two AI components, along with an ICT-developed perception system sitting atop the Gamebryo rendering engine. The typical number of agents supported on a dual-core CPU with a modern graphics card is ~10,000 rendering at 30 frames-per-second. To support this quantity and level of intelligence several design considerations were implemented, including the use of multiple threads, a clone/sprite-based avatar view, and a dynamic level-of-detail update system. Future work includes distributing the AI mechanism across multiple machines to support numbers of agents a level of magnitude higher than is currently possible.