Learn More
The Supply Chain Management track of the international Trading Agents Competition (TAC SCM) was introduced in 2003 as a test-bed for researchers interested in building autonomous agents that act in dynamic supply chains. TAC SCM provides a challenging scenario for existing AI decision-making algorithms, due to the high dimensionality and the non-determinism(More)
We address the problem of automatically constructing basis functions for linear approximation of the value function of a Markov Decision Process (MDP). Our work builds on results by Bertsekas and Castañon (1989) who proposed a method for automatically aggregating states to speed up value iteration. We propose to use neighborhood component analysis(More)
The Supply Chain Management track of the international trading Agents Competition (TAC ACM) provides a challenging scenario for existing AI decision-making algorithms, due to the high dimensionality and the non-determinism of the environment. Our entry in this competition, RedAgent, is centered around the idea of using an internal market in order to(More)
The standard approach for learning Markov Models with Hidden State uses the Expectation-Maximization framework. While this approach had a significant impact on several practical applications (e.g. speech recognition, biological sequence alignment) it has two major limitations: it requires a known model topology, and learning is only locally optimal. We(More)
The truthmaker theory rests on the thesis that the link between a true judgment and that in the world to which it corresponds is not a one-to-one but rather a one-to-many relation. An analogous thesis in relation to the link between a singular term and that in the world to which it refers is already widely accepted. This is the thesis to the effect that(More)
  • 1