In this paper we consider the broader issue of gaining assurance that an agent system will behave appropriately when it is deployed. We ask to what extent this problem is addressed by existing research into formal verification. We identify a range of issues with existing work which leads us to conclude that, broadly speaking, verification approaches on their own are too narrowly focussed. We argue that a shift in direction is needed, and outline some possibilities for such a shift in direction.
Download (PDF, 296 KB)
Novelty detection is an important functionality that has found many applications in information retrieval and processing. In this paper we propose a novel framework that deals with novelty detection for multiple-scene image sets. Working with wildlife image data, the framework starts with image segmentation, followed by feature extraction and classification of the image blocks extracted from image segments. The labelled image blocks are then scanned through to generate a co-occurrence matrix of object labels, representing the semantic context within the scene. The semantic co-occurrence matrices then undergo binarization and principal component analysis for dimension reduction, forming the basis for constructing one-class models for each scene category. An algorithm for outlier detection that employs multiple one-class models is proposed. An advantage of our approach is that it can be used for scene classification and novelty detection at the same time. Our experiments show that the proposed approach algorithm gives favourable performance for the task of detecting novel wildlife scenes, and binarization of the label co-occurrence matrices helps to significantly increase the robustness in dealing with the variation of scene statistics.
Keywords: context, co-occurrence matrix, semantics, novel image, multi-class
Corrected version uploaded 2010-05-04.
Download (PDF, 3.4 MB)
In normative multi-agent systems, the question of “how an agent identifies a norm in an agent society” has not received much attention. This paper aims at addressing this question. To this end, this paper proposes an architecture for norm identification for an agent. The architecture is based on observation of interactions between agents. This architecture enables an autonomous agent to identify the norms in a society using the Candidate Norm Inference (CNI) algorithm. The CNI algorithm uses association rule mining approach to identify sequences of events as candidate norms. When a norm changes, the agent using our architecture will be able to modify the norm and also remove a norm if it does not hold in its society. Using simulations we demonstrate how an agent makes use of the norm identification framework.
Keywords: norms, agents, architecture, identification, simulation, societies
Download (PDF, 562 KB)
Modeling of financial market data for detecting important market characteristics as well as their abnormalities plays a key role in identifying their behavior. Researchers have proposed different types of techniques to model market data. One such model proposed by Sergie Maslov, models the behavior of a limit order book. Being a very simple and interesting model, it has several drawbacks and limitations.
This paper analyses the behavior of the Maslov model and proposes several variants of it to make the original Maslov model more realistic. The price signals generated from these models are analyzed by comparing with real life stock data and it was shown that the proposed variants of the Maslov model are more realistic than the original Maslov model.
Download (PDF, 832 KB)