On the Relationship Between the Generalized Equality Classifier and ART 2 Neural Networks

In this paper, we introduce the Generalized Equality Classifier (GEC) for use as an unsupervised clustering algorithm in categorizing analog data. GEC is based on a formal definition of inexact equality originally developed for voting in fault tolerant software applications. GEC is defined using a metric space framework. The only parameter in GEC is a scalar threshold which defines the approximate equality of two patterns. Here, we compare the characteristics of GEC to the ART2-A algorithm (Carpenter, Grossberg, and Rosen, 1991). In particular, we show that GEC with the Hamming distance performs the same optimization as ART2. Moreover, GEC has lower computational requirements than AR12 on serial machines.

Integration of relational and hierarchical network information for protein function prediction

BACKGROUND:In the current climate of high-throughput computational biology, the inference of a protein's function from related measurements, such as protein-protein interaction relations, has become a canonical task. Most existing technologies pursue this task as a classification problem, on a term-by-term basis, for each term in a database, such as the Gene Ontology (GO) database, a popular rigorous vocabulary for biological functions. However, ontology structures are essentially hierarchies, with certain top to bottom annotation rules which protein function predictions should in principle follow. Currently, the most common approach to imposing these hierarchical constraints on network-based classifiers is through the use of transitive closure to predictions.RESULTS:We propose a probabilistic framework to integrate information in relational data, in the form of a protein-protein interaction network, and a hierarchically structured database of terms, in the form of the GO database, for the purpose of protein function prediction. At the heart of our framework is a factorization of local neighborhood information in the protein-protein interaction network across successive ancestral terms in the GO hierarchy. We introduce a classifier within this framework, with computationally efficient implementation, that produces GO-term predictions that naturally obey a hierarchical 'true-path' consistency from root to leaves, without the need for further post-processing.CONCLUSION:A cross-validation study, using data from the yeast Saccharomyces cerevisiae, shows our method offers substantial improvements over both standard 'guilt-by-association' (i.e., Nearest-Neighbor) and more refined Markov random field methods, whether in their original form or when post-processed to artificially impose 'true-path' consistency. Further analysis of the results indicates that these improvements are associated with increased predictive capabilities (i.e., increased positive predictive value), and that this increase is consistent uniformly with GO-term depth. Additional in silico validation on a collection of new annotations recently added to GO confirms the advantages suggested by the cross-validation study. Taken as a whole, our results show that a hierarchical approach to network-based protein function prediction, that exploits the ontological structure of protein annotation databases in a principled manner, can offer substantial advantages over the successive application of 'flat' network-based methods.

A Bayesian Approach for TCP to Distinguish Congestion from Wireless Losses

(This Technical Report revises TR-BUCS-2003-011) The Transmission Control Protocol (TCP) has been the protocol of choice for many Internet applications requiring reliable connections. The design of TCP has been challenged by the extension of connections over wireless links. In this paper, we investigate a Bayesian approach to infer at the source host the reason of a packet loss, whether congestion or wireless transmission error. Our approach is "mostly" end-to-end since it requires only one long-term average quantity (namely, long-term average packet loss probability over the wireless segment) that may be best obtained with help from the network (e.g. wireless access agent).Specifically, we use Maximum Likelihood Ratio tests to evaluate TCP as a classifier of the type of packet loss. We study the effectiveness of short-term classification of packet errors (congestion vs. wireless), given stationary prior error probabilities and distributions of packet delays conditioned on the type of packet loss (measured over a larger time scale). Using our Bayesian-based approach and extensive simulations, we demonstrate that congestion-induced losses and losses due to wireless transmission errors produce sufficiently different statistics upon which an efficient online error classifier can be built. We introduce a simple queueing model to underline the conditional delay distributions arising from different kinds of packet losses over a heterogeneous wired/wireless path. We show how Hidden Markov Models (HMMs) can be used by a TCP connection to infer efficiently conditional delay distributions. We demonstrate how estimation accuracy is influenced by different proportions of congestion versus wireless losses and penalties on incorrect classification.

Automatic 2D Hand Tracking in Video Sequences

In gesture and sign language video sequences, hand motion tends to be rapid, and hands frequently appear in front of each other or in front of the face. Thus, hand location is often ambiguous, and naive color-based hand tracking is insufficient. To improve tracking accuracy, some methods employ a prediction-update framework, but such methods require careful initialization of model parameters, and tend to drift and lose track in extended sequences. In this paper, a temporal filtering framework for hand tracking is proposed that can initialize and reset itself without human intervention. In each frame, simple features like color and motion residue are exploited to identify multiple candidate hand locations. The temporal filter then uses the Viterbi algorithm to select among the candidates from frame to frame. The resulting tracking system can automatically identify video trajectories of unambiguous hand motion, and detect frames where tracking becomes ambiguous because of occlusions or overlaps. Experiments on video sequences of several hundred frames in duration demonstrate the system's ability to track hands robustly, to detect and handle tracking ambiguities, and to extract the trajectories of unambiguous hand motion.

Indexing Methods for Efficient Multiclass Recognition

Many real world image analysis problems, such as face recognition and hand pose estimation, involve recognizing a large number of classes of objects or shapes. Large margin methods, such as AdaBoost and Support Vector Machines (SVMs), often provide competitive accuracy rates, but at the cost of evaluating a large number of binary classifiers, thus making it difficult to apply such methods when thousands or millions of classes need to be recognized. This thesis proposes a filter-and-refine framework, whereby, given a test pattern, a small number of candidate classes can be identified efficiently at the filter step, and computationally expensive large margin classifiers are used to evaluate these candidates at the refine step. Two different filtering methods are proposed, ClassMap and OVA-VS (One-vs.-All classification using Vector Search). ClassMap is an embedding-based method, works for both boosted classifiers and SVMs, and tends to map the patterns and their associated classes close to each other in a vector space. OVA-VS maps OVA classifiers and test patterns to vectors based on the weights and outputs of weak classifiers of the boosting scheme. At runtime, finding the strongest-responding OVA classifier becomes a classical vector search problem, where well-known methods can be used to gain efficiency. In our experiments, the proposed methods achieve significant speed-ups, in some cases up to two orders of magnitude, compared to exhaustive evaluation of all OVA classifiers. This was achieved in hand pose recognition and face recognition systems where the number of classes ranges from 535 to 48,600.

Forewarding in Mobile Opportunistic Networks

Recent advances in processor speeds, mobile communications and battery life have enabled computers to evolve from completely wired to completely mobile. In the most extreme case, all nodes are mobile and communication takes place at available opportunities – using both traditional communication infrastructure as well as the mobility of intermediate nodes. These are mobile opportunistic networks. Data communication in such networks is a difficult problem, because of the dynamic underlying topology, the scarcity of network resources and the lack of global information. Establishing end-to-end routes in such networks is usually not feasible. Instead a store-and-carry forwarding paradigm is better suited for such networks. This dissertation describes and analyzes algorithms for forwarding of messages in such networks. In order to design effective forwarding algorithms for mobile opportunistic networks, we start by first building an understanding of the set of all paths between nodes, which represent the available opportunities for any forwarding algorithm. Relying on real measurements, we enumerate paths between nodes and uncover what we refer to as the path explosion effect. The term path explosion refers to the fact that the number of paths between a randomly selected pair of nodes increases exponentially with time. We draw from the theory of epidemics to model and explain the path explosion effect. This is the first contribution of the thesis, and is a key observation that underlies subsequent results. Our second contribution is the study of forwarding algorithms. For this, we rely on trace driven simulations of different algorithms that span a range of design dimensions. We compare the performance (success rate and average delay) of these algorithms. We make the surprising observation that most algorithms we consider have roughly similar performance. We explain this result in light of the path explosion phenomenon. While the performance of most algorithms we studied was roughly the same, these algorithms differed in terms of cost. This prompted us to focus on designing algorithms with the explicit intent of reducing costs. For this, we cast the problem of forwarding as an optimal stopping problem. Our third main contribution is the design of strategies based on optimal stopping principles which we refer to as Delegation schemes. Our analysis shows that using a delegation scheme reduces cost over naive forwarding by a factor of O(√N), where N is the number of nodes in the network. We further validate this result on real traces, where the cost reduction observed is even greater. Our results so far include a key assumption, which is unbounded buffers on nodes. Next, we relax this assumption, so that the problem shifts to one of prioritization of messages for transmission and dropping. Our fourth contribution is the study of message prioritization schemes, combined with forwarding. Our main result is that one achieves higher performance by assigning higher priorities to young messages in the network. We again interpret this result in light of the path explosion effect.

An alignment based similarity measure for hand detection in cluttered sign language video

Locating hands in sign language video is challenging due to a number of factors. Hand appearance varies widely across signers due to anthropometric variations and varying levels of signer proficiency. Video can be captured under varying illumination, camera resolutions, and levels of scene clutter, e.g., high-res video captured in a studio vs. low-res video gathered by a web cam in a user’s home. Moreover, the signers’ clothing varies, e.g., skin-toned clothing vs. contrasting clothing, short-sleeved vs. long-sleeved shirts, etc. In this work, the hand detection problem is addressed in an appearance matching framework. The Histogram of Oriented Gradient (HOG) based matching score function is reformulated to allow non-rigid alignment between pairs of images to account for hand shape variation. The resulting alignment score is used within a Support Vector Machine hand/not-hand classifier for hand detection. The new matching score function yields improved performance (in ROC area and hand detection rate) over the Vocabulary Guided Pyramid Match Kernel (VGPMK) and the traditional, rigid HOG distance on American Sign Language video gestured by expert signers. The proposed match score function is computationally less expensive (for training and testing), has fewer parameters and is less sensitive to parameter settings than VGPMK. The proposed detector works well on test sequences from an inexpert signer in a non-studio setting with cluttered background.

3D Trajectory Recovery for Tracking Multiple Objects and Trajectory Guided Recognition of Actions

A mechanism is proposed that integrates low-level (image processing), mid-level (recursive 3D trajectory estimation), and high-level (action recognition) processes. It is assumed that the system observes multiple moving objects via a single, uncalibrated video camera. A novel extended Kalman filter formulation is used in estimating the relative 3D motion trajectories up to a scale factor. The recursive estimation process provides a prediction and error measure that is exploited in higher-level stages of action recognition. Conversely, higher-level mechanisms provide feedback that allows the system to reliably segment and maintain the tracking of moving objects before, during, and after occlusion. The 3D trajectory, occlusion, and segmentation information are utilized in extracting stabilized views of the moving object. Trajectory-guided recognition (TGR) is proposed as a new and efficient method for adaptive classification of action. The TGR approach is demonstrated using "motion history images" that are then recognized via a mixture of Gaussian classifier. The system was tested in recognizing various dynamic human outdoor activities; e.g., running, walking, roller blading, and cycling. Experiments with synthetic data sets are used to evaluate stability of the trajectory estimator with respect to noise.

Trajectory Guided Tracking and Recognition of Actions

A combined 2D, 3D approach is presented that allows for robust tracking of moving people and recognition of actions. It is assumed that the system observes multiple moving objects via a single, uncalibrated video camera. Low-level features are often insufficient for detection, segmentation, and tracking of non-rigid moving objects. Therefore, an improved mechanism is proposed that integrates low-level (image processing), mid-level (recursive 3D trajectory estimation), and high-level (action recognition) processes. A novel extended Kalman filter formulation is used in estimating the relative 3D motion trajectories up to a scale factor. The recursive estimation process provides a prediction and error measure that is exploited in higher-level stages of action recognition. Conversely, higher-level mechanisms provide feedback that allows the system to reliably segment and maintain the tracking of moving objects before, during, and after occlusion. The 3D trajectory, occlusion, and segmentation information are utilized in extracting stabilized views of the moving object that are then used as input to action recognition modules. Trajectory-guided recognition (TGR) is proposed as a new and efficient method for adaptive classification of action. The TGR approach is demonstrated using "motion history images" that are then recognized via a mixture-of-Gaussians classifier. The system was tested in recognizing various dynamic human outdoor activities: running, walking, roller blading, and cycling. Experiments with real and synthetic data sets are used to evaluate stability of the trajectory estimator with respect to noise.

Efficient Hash-Consing of Recursive Types

Efficient storage of types within a compiler is necessary to avoid large blowups in space during compilation. Recursive types in particular are important to consider, as naive representations of recursive types may be arbitrarily larger than necessary through unfolding. Hash-consing has been used to efficiently store non-recursive types. Deterministic finite automata techniques have been used to efficiently perform various operations on recursive types. We present a new system for storing recursive types combining hash-consing and deterministic finite automata techniques. The space requirements are linear in the number of distinct types. Both update and lookup operations take polynomial time and linear space and type equality can be checked in constant time once both types are in the system.

SymbolDesign: A User-centered Method to Design Pen-based Interfaces and Extend the Functionality of Pointer Input Devices

A method called "SymbolDesign" is proposed that can be used to design user-centered interfaces for pen-based input devices. It can also extend the functionality of pointer input devices such as the traditional computer mouse or the Camera Mouse, a camera-based computer interface. Users can create their own interfaces by choosing single-stroke movement patterns that are convenient to draw with the selected input device and by mapping them to a desired set of commands. A pattern could be the trace of a moving finger detected with the Camera Mouse or a symbol drawn with an optical pen. The core of the SymbolDesign system is a dynamically created classifier, in the current implementation an artificial neural network. The architecture of the neural network automatically adjusts according to the complexity of the classification task. In experiments, subjects used the SymbolDesign method to design and test the interfaces they created, for example, to browse the web. The experiments demonstrated good recognition accuracy and responsiveness of the user interfaces. The method provided an easily-designed and easily-used computer input mechanism for people without physical limitations, and, with some modifications, has the potential to become a computer access tool for people with severe paralysis.

Efficient Nearest Neighbor Classification Using a Cascade of Approximate Similarity Measures

Nearest neighbor classification using shape context can yield highly accurate results in a number of recognition problems. Unfortunately, the approach can be too slow for practical applications, and thus approximation strategies are needed to make shape context practical. This paper proposes a method for efficient and accurate nearest neighbor classification in non-Euclidean spaces, such as the space induced by the shape context measure. First, a method is introduced for constructing a Euclidean embedding that is optimized for nearest neighbor classification accuracy. Using that embedding, multiple approximations of the underlying non-Euclidean similarity measure are obtained, at different levels of accuracy and efficiency. The approximations are automatically combined to form a cascade classifier, which applies the slower approximations only to the hardest cases. Unlike typical cascade-of-classifiers approaches, that are applied to binary classification problems, our method constructs a cascade for a multiclass problem. Experiments with a standard shape data set indicate that a two-to-three order of magnitude speed up is gained over the standard shape context classifier, with minimal losses in classification accuracy.

Face Identification by a Cascade of Rejection Classifiers

Nearest neighbor search is commonly employed in face recognition but it does not scale well to large dataset sizes. A strategy to combine rejection classifiers into a cascade for face identification is proposed in this paper. A rejection classifier for a pair of classes is defined to reject at least one of the classes with high confidence. These rejection classifiers are able to share discriminants in feature space and at the same time have high confidence in the rejection decision. In the face identification problem, it is possible that a pair of known individual faces are very dissimilar. It is very unlikely that both of them are close to an unknown face in the feature space. Hence, only one of them needs to be considered. Using a cascade structure of rejection classifiers, the scope of nearest neighbor search can be reduced significantly. Experiments on Face Recognition Grand Challenge (FRGC) version 1 data demonstrate that the proposed method achieves significant speed up and an accuracy comparable with the brute force Nearest Neighbor method. In addition, a graph cut based clustering technique is employed to demonstrate that the pairwise separability of these rejection classifiers is capable of semantic grouping.

Web Mediators for Accessible Browsing

We present a highly accurate method for classifying web pages based on link percentage, which is the percentage of text characters that are parts of links normalized by the number of all text characters on a web page. K-means clustering is used to create unique thresholds to differentiate index pages and article pages on individual web sites. Index pages contain mostly links to articles and other indices, while article pages contain mostly text. We also present a novel link grouping algorithm using agglomerative hierarchical clustering that groups links in the same spatial neighborhood together while preserving link structure. Grouping allows users with severe disabilities to use a scan-based mechanism to tab through a web page and select items. In experiments, we saw up to a 40-fold reduction in the number of commands needed to click on a link with a scan-based interface, which shows that we can vastly improve the rate of communication for users with disabilities. We used web page classification and link grouping to alter web page display on an accessible web browser that we developed to make a usable browsing interface for users with disabilities. Our classification method consistently outperformed a baseline classifier even when using minimal data to generate article and index clusters, and achieved classification accuracy of 94.0% on web sites with well-formed or slightly malformed HTML, compared with 80.1% accuracy for the baseline classifier.

Implications of Selfish Neighbor Selection in Overlay Networks

In a typical overlay network for routing or content sharing, each node must select a fixed number of immediate overlay neighbors for routing traffic or content queries. A selfish node entering such a network would select neighbors so as to minimize the weighted sum of expected access costs to all its destinations. Previous work on selfish neighbor selection has built intuition with simple models where edges are undirected, access costs are modeled by hop-counts, and nodes have potentially unbounded degrees. However, in practice, important constraints not captured by these models lead to richer games with substantively and fundamentally different outcomes. Our work models neighbor selection as a game involving directed links, constraints on the number of allowed neighbors, and costs reflecting both network latency and node preference. We express a node's "best response" wiring strategy as a k-median problem on asymmetric distance, and use this formulation to obtain pure Nash equilibria. We experimentally examine the properties of such stable wirings on synthetic topologies, as well as on real topologies and maps constructed from PlanetLab and AS-level Internet measurements. Our results indicate that selfish nodes can reap substantial performance benefits when connecting to overlay networks composed of non-selfish nodes. On the other hand, in overlays that are dominated by selfish nodes, the resulting stable wirings are optimized to such great extent that even non-selfish newcomers can extract near-optimal performance through naive wiring strategies.