Goal-conditioned policies are generally understood to be "feed-forward" circuits,

Goal-conditioned policies are generally understood to be "feed-forward" circuits, in the form of neural networks that map from the current state and the goal specification to the next action to take. However, under what circumstances such a policy can be learned and how efficient the policy will be are not well understood. In this paper, we present a circuit complexity analysis for relational neural networks (such as graph neural networks and transformers) representing policies for planning problems, by drawing connections with serialized goal regression search (S-GRS). We show that there are three general classes of planning problems, in terms of the growth of circuit width and depth as a function of the number of objects and planning horizon, providing constructive proofs. We also illustrate the utility of this analysis for designing neural networks for policy learning.

Recently, Qiao, Duan, and Cheng~(2019) proposed a distributed nearest-neighbor classification method, in which a massive dataset is split into smaller groups, each processed with a $k$-nearest-neighbor classifier, and the final class label is predicted by a majority vote among these groupwise class labels. This paper shows that the distributed algorithm with $k=1$ over a sufficiently large number of groups attains a minimax optimal error rate up to a multiplicative logarithmic factor under some regularity conditions, for both regression and classification problems. Roughly speaking, distributed 1-nearest-neighbor rules with $M$ groups has a performance comparable to standard $\Theta(M)$-nearest-neighbor rules. In the analysis, alternative rules with a refined aggregation method are proposed and shown to attain exact minimax optimal rates.

Many real world problems can be defined as optimisation problems in which the aim is to maximise an objective function. The quality of obtained solution is directly linked to the pertinence of the used objective function. However, designing such function, which has to translate the user needs, is usually fastidious. In this paper, a method to help user objective functions designing is proposed. Our approach, which is highly interactive, is based on man machine dialogue and more particularly on the comparison of problem instance solutions by the user. We propose an experiment in the domain of cartographic generalisation that shows promising results.

In this paper, we propose a novel approach for learning regression rules by transforming the regression problem into a classification problem. Unlike previous approaches to regression by classification, in our approach the discretization of the class variable is tightly integrated into the rule learning algorithm. The key idea is to dynamically define a region around the target value predicted by the rule, and considering all examples within that region as positive and all examples outside that region as negative. In this way, conventional rule learning heuristics may be used for inducing regression rules. Our results show that our heuristic algorithm outperforms approaches that use a static discretization of the target variable, and performs en par with other comparable rule-based approaches, albeit without reaching the performance of statistical approaches.

Reprinted as Chapter 3 (pp.250-271) of Webber & Nilsson, Readings in Artificial IntelligenceElcock, E. W. and Michie, D. (Eds.), Machine Intelligence 8, pp. 94-€“138. Ellis Horwood.