An important issue in the application of machine learning techniques to information management tasks is the nature of features extracted from textual information. We have created an intelligent email agent that can learn actions such as filtering, prioritizing, downloading to palmtops, and forwarding email to voicemail using automatic feature extraction. Our agent's newfeature extraction approach is based on first learning concepts present within the mail, then using these concepts as features for learning actions to perform on the messages. What features should be chosen? This paper describes the concept features approach and considers two sources for learning conceptual features: groups defined by the user and groups defined by the agent's task. Additionally, features may be defined by vectorized examples or keywords. Experimental results are provided for an email sorting task. Keywords Electronic mail, intelligent agents, machine learning, information management, feature selection...

Abstract not found

Abstract — This paper presents a tutorial overview of neural networks as signal processing tools for image compression. They are well suited to the problem of image compression due to their massively parallel and distributed architecture. Their characteristics are analogous to some of the features of our own visual system, which allow us to process visual information with much ease. For example, multilayer perceptrons can be used as nonlinear predictors in differential pulse-code modulation (DPCM). Such predictors have been shown to increase the predictive gain relative to a linear predictor. Another active area of research is in the application of Hebbian learning to the extraction of principal components, which are the basis vectors for the optimal linear Karhunen-Loève transform (KLT). These learning algorithms are iterative, have some computational advantages over standard eigendecomposition techniques, and can be made to adapt to changes in the input signal. Yet another model, the self-organizing feature map (SOFM), has been used with a great deal of success in the design of codebooks for vector quantization (VQ). The resulting codebooks are less sensitive to initial conditions than the standard LBG algorithm, and the topological ordering of the entries can be exploited to further increase coding efficiency and reduce computational complexity. I.

Spatiotemporal connectionist networks (STCN's) comprise an important class of neural models that can deal with patterns distributed both in time and space. In this paper, we widen the application domain of the taxonomy for supervised STCN's recently proposed by Kremer (2001) to the unsupervised case. This is possible through a reinterpretation of the state vector as a vector of latent (hidden) variables, as proposed by Meinicke (2000). The goal of this generalized taxonomy is then to provide a nonlinear generative framework for describing unsupervised spatiotemporal networks, making it easier to compare and contrast their representational and operational characteristics. Computational properties, representational issues and learning are also discussed and a number of references to the relevant source publications are provided. It is argued that the proposed approach is simple and more powerful than the previous attempts, from a descriptive and predictive viewpoint. We also discuss the relation of this taxonomy with automata theory and state space modeling, and suggest directions for further work.

Abstract: Adaptively weighted Sub-pattern PCA (Aw-SpPCA) for face recognition is presented in this paper. Unlike PCA based on a whole image pattern, Aw-SpPCA operates directly on its sub-patterns partitioned from an original whole pattern and separately extracts features from them. Moreover, unlike both SpPCA and mPCA that neglect different contributions made by different parts of the human face in face recognition, Aw-SpPCA can adaptively compute the contributions of each part and then endows them to a classification task in order to enhance the robustness to both expression and illumination variations. Experiments on three standard face databases show that the proposed method is competitive.

visual structure: human data and a model

When interpreting the results of experiments that investigate biological development, one is faced with a wealth of data. Producing a model of such development must always involve some degree of abstraction. The appropriate level of abstraction and the importance of particular experimental evidence is determined by one's modelling objective. Models may potentially be motivated by one of two complementary aims: 1. To understand how biological neurons achieve their mature interconnectivity...

This report centres around some of this issues involved in automatic transcription of polyphonic musical audio signals. That is, representing the information contained in the audio in such a way as to be recognisable and usable by a musician. First, a review of the various fields which have a bearing on the subject is put forward, including music, music psychology, auditory psychology and signal processing. Then a thorough appraisal of previous work on automated polyphonic transcription is presented. Next, original work on the use of time-frequency reassignment as a front end is imparted and finally, future ideas are expounded and a timetable for forthcoming research is given.

Copyright c 1997 R.O. Rogers and D.B. Skillicorn We use the cost system of BSP (Bulk Synchronous Parallelism) to predict the performance of three di erent parallelization techniques for both supervised and unsupervised learning in arti cial neural networks. We show that exemplar parallelism outperforms techniques that partition the neural network across processors, especially when the number of exemplars is large, typical of applications such as data mining.

Abstract not found