The information rate of finite-state source/channel models can be accurately estimated by sampling both a long channel input sequence and the corresponding channel output sequence, followed by a forward sum–product recursion on the joint source/channel trellis. This method is extended to compute upper and lower bounds on the information rate of very general channels with memory by means of finite-state approximations. Further upper and lower bounds can be computed by reduced-state methods.

Modeling dynamic human control strategy (HCS), or human skill in response to real-time sensing is becoming an increasingly popular paradigm in many different research areas, such as intelligent vehicle systems, virtual reality, and space robotics. Such models are often learned from experimental data, and as such can be characterized despite the lack of a good physical model. Unfortunately, learned models presently offer few, if any, guarantees in terms of model fidelity to the training data. This is especially true for dynamic reaction skills, where errors can feed back on themselves to generate state and command trajectories uncharacteristic of the source process. Thus, we propose a stochastic similarity measure---based on hidden Markov model (HMM) analysis---capable of comparing and contrasting stochastic, dynamic, multidimensional trajectories. This similarity measure is the first step in validating a learned model's fidelity to its training data by comparing the model's dynamic trajectories in the feedback loop to the human's dynamic trajectories. In this paper, we first derive and demonstrate properties of the similarity measure for stochastic systems. We then apply the similarity measure to real-time human driving data by comparing different control strategies among different individuals. We show that the proposed similarity measure out performs the more traditional Bayes classifier in correctly grouping driving data from the same individual. Finally, we illustrate how the similarity measure can be used in the validation of models which are learned from experimental data, and how we can connect model validation and model learning to iteratively improve our models of human control strategy.

The emergence of digital music on the Internet requires new information retrieval methods adapted to the specific characteristics and needs. While music retrieval based on the text information, such as title, composers, or subject classification, has been implemented in many existing systems, retrieval of a piece of music based on music contents, especially based on an incomplete, imperfect recall of a fragment of the music, has not yet been fully explored. This thesis is to explore the main problems involved in a web-based melody retrieval system. I propose to build a query-by-humming system, which can find a piece of music in the digital music repository based on a few hummed notes, using a melody representation that combines both the pitch contour and the beat information. Since an input query (hummed melody) may have various errors due to uncertainty of the user's memory or the user's singing ability, the system should be able to tolerate the errors. Furthermore, extracting m...

Abstract. This paper describes a clustering methodology for temporal data using hidden Markov model(HMM) representation. The proposed method improves upon existing HMM based clustering methods in two ways: (i) it enables HMMs to dynamically change its model structure to obtain a better t model for data during clustering process, and (ii) it provides objective criterion function to automatically select the clustering partition. The algorithm is presented in terms of four nested levels of searches: (i) the search for the number of clusters in a partition, (ii) the search for the structure for a xed sized partition, (iii) the search for the HMM structure for each cluster, and (iv) the search for the parameter values for each HMM. Preliminary experiments with arti cially generated data demonstrate the e ectiveness of the proposed methodology. 1

This research aims to support collaborative distance learners by demonstrating how a probabilistic machine learning method can be used to model and analyze online knowledge sharing interactions. The approach applies Hidden Markov Models and Multidimensional Scaling to analyze and assess sequences of coded online student interaction. These analysis techniques were used to train a system to dynamically recognize (1) when students are having trouble learning the new concepts they share with each other, and (2) why they are having trouble. The results of this research may assist an instructor or intelligent coach in understanding and mediating situations in which groups of students collaborate to share their knowledge.

We propose a novel similarity measure for Hidden Markov Models (HMMs). This measure calculates the Bayes probability of error for HMM state correspondences and propagates it along the Viterbi path in a similar way to the HMM Viterbi scoring. It can be applied as a tool to interpret misclassifications, as a stop criterion in iterative HMM training or as a distance measure for HMM clustering. The similarity measure is evaluated in the context of online handwriting recognition on lower case character models which have been trained from the UNIPEN database. We compare the similarities with experimental classifications. The results show that similar and misclassified class pairs are highly correlated. The measure is not limited to handwriting recognition, but can be used in other applications that use HMM based methods.

We present an approach to cluster the training data for automatic speech recognition (ASR). A relativeentropy based distance metric between training data clusters is defined. This metric is used to hierarchically cluster the training data. The metric can also be used to select the closest training data clusters given a small amount of data from the test speaker. The selected clusters are then used to estimate a set of hidden Markov models (HMMs) for recognizing the speech from the test speaker. We present preliminary experimental results of the clustering algorithm and its application to ASR. 1 Introduction While progress in ASR has been encouraging, it has become increasingly clear that ASR systems must perform well in the presence of mismatches between the training and testing environments. ASR systems trained in one environment often perform poorly in a new environment due to mismatches between the training and testing conditions. Common sources of mismatches include different tran...

Hidden Markov models (HMMs)and partially observable Markov decision processes (POMDPs) form a useful tool for modeling dynamical systems. They are particularly useful for representing environments such as road networks and office buildings, which are typical for robot navigation and planning. The work presented here is concerned with acquiring such models. We demonstrate how domain-specific information and constraints can be incorporated into the statistical estimation process, greatly improving the learned models in terms of the model quality, the number of iterations required for convergence and robustness to reduction in the amount of available data. We present new initialization heuristics which can be used even when the data suffers from cumulative rotational error, new update rules for the model parameters, as an instance of generalized EM, and a strategy for enforcing complete geometrical consistency in the model. Experimental results demonstrate the effectiveness of our approac...

We explain how to apply statistical techniques to solve several language-recognition problems that arise in cryptanalysis and other domains. Language recognition is important in cryptanalysis because, among other applications, an exhaustive key search of any cryptosystem from ciphertext alone requires a test that recognizes valid plaintext. Written for cryptanalysts, this guide should also be helpful to others as an introduction to statistical inference on Markov chains. Modeling language as a finite stationary Markov process, we adapt a statistical model of pattern recognition to language recognition. Within this framework we consider four welldefined language-recognition problems: 1) recognizing a known language, 2) distinguishing a known language from uniform noise, 3) distinguishing unknown 0th-order noise from unknown 1st-order language, and 4) detecting non-uniform unknown language. For the second problem we give a most powerful test based on the Neyman-Pearson Lemma. For the oth...

This paper investigates two methods to define a distance measure between any pair of Hidden Markov Models (HMM). The first one is the geometricaly motivated euclidean distance which solely incorporates the feature probabilities. The second mesures is the Kulback-Liebler distance which is based on the discriminating power of the probability measure on the space of feature sequences induced by the HMMs. A method is shown, to compute the proposed measures reasonable fast and the distance measures are compared in a series of simulations involving HMMs from a real world speech recognition system. 1. INTRODUCTION Hidden Markov Models have been applied in various research fields. Their success is mainly based upon the existence of a automatic iterativ learning algorithm [1,6] which adjusts the parameter of a HMM to a given training sequence. However, there is no canonical way to measure the dissimilarity between two different HMM. The need for such a distance measure arises in an automatic s...