RICH AND EFFICIENT VISUAL DATA REPRESENTATION

Increasing the size of training data in many computer vision tasks has shown to be very effective. Using large scale image datasets (e.g. ImageNet) with simple learning techniques (e.g. linear classifiers) one can achieve state-of-the-art performance in object recognition compared to sophisticated learning techniques on smaller image sets. Semantic search on visual data has become very popular. There are billions of images on the internet and the number is increasing every day. Dealing with large scale image sets is intense per se. They take a significant amount of memory that makes it impossible to process the images with complex algorithms on single CPU machines. Finding an efficient image representation can be a key to attack this problem. A representation being efficient is not enough for image understanding. It should be comprehensive and rich in carrying semantic information. In this proposal we develop an approach to computing binary codes that provide a rich and efficient image representation. We demonstrate several tasks in which binary features can be very effective. We show how binary features can speed up large scale image classification. We present learning techniques to learn the binary features from supervised image set (With different types of semantic supervision; class labels, textual descriptions). We propose several problems that are very important in finding and using efficient image representation.

User Behavioral Modeling of Web-based Systems for Continuous User Authentication

Authentication plays an important role in how we interact with computers, mobile devices, the web, etc. The idea of authentication is to uniquely identify a user before granting access to system privileges. For example, in recent years more corporate information and applications have been accessible via the Internet and Intranet. Many employees are working from remote locations and need access to secure corporate files. During this time, it is possible for malicious or unauthorized users to gain access to the system. For this reason, it is logical to have some mechanism in place to detect whether the logged-in user is the same user in control of the user's session. Therefore, highly secure authentication methods must be used. We posit that each of us is unique in our use of computer systems. It is this uniqueness that is leveraged to "continuously authenticate users" while they use web software. To monitor user behavior, n-gram models are used to capture user interactions with web-based software. This statistical language model essentially captures sequences and sub-sequences of user actions, their orderings, and temporal relationships that make them unique by providing a model of how each user typically behaves. Users are then continuously monitored during software operations. Large deviations from "normal behavior" can possibly indicate malicious or unintended behavior. This approach is implemented in a system called Intruder Detector (ID) that models user actions as embodied in web logs generated in response to a user's actions. User identification through web logs is cost-effective and non-intrusive. We perform experiments on a large fielded system with web logs of approximately 4000 users. For these experiments, we use two classification techniques; binary and multi-class classification. We evaluate model-specific differences of user behavior based on coarse-grain (i.e., role) and fine-grain (i.e., individual) analysis. A specific set of metrics are used to provide valuable insight into how each model performs. Intruder Detector achieves accurate results when identifying legitimate users and user types. This tool is also able to detect outliers in role-based user behavior with optimal performance. In addition to web applications, this continuous monitoring technique can be used with other user-based systems such as mobile devices and the analysis of network traffic.

Wave impedance selection for passivity-based bilateral teleoperation

When a task must be executed in a remote or dangerous environment, teleoperation systems may be employed to extend the influence of the human operator. In the case of manipulation tasks, haptic feedback of the forces experienced by the remote (slave) system is often highly useful in improving an operator's ability to perform effectively. In many of these cases (especially teleoperation over the internet and ground-to-space teleoperation), substantial communication latency exists in the control loop and has the strong tendency to cause instability of the system. The first viable solution to this problem in the literature was based on a scattering/wave transformation from transmission line theory. This wave transformation requires the designer to select a wave impedance parameter appropriate to the teleoperation system. It is widely recognized that a small value of wave impedance is well suited to free motion and a large value is preferable for contact tasks. Beyond this basic observation, however, very little guidance exists in the literature regarding the selection of an appropriate value. Moreover, prior research on impedance selection generally fails to account for the fact that in any realistic contact task there will simultaneously exist contact considerations (perpendicular to the surface of contact) and quasi-free-motion considerations (parallel to the surface of contact). The primary contribution of the present work is to introduce an approximate linearized optimum for the choice of wave impedance and to apply this quasi-optimal choice to the Cartesian reality of such a contact task, in which it cannot be expected that a given joint will be either perfectly normal to or perfectly parallel to the motion constraint. The proposed scheme selects a wave impedance matrix that is appropriate to the conditions encountered by the manipulator. This choice may be implemented as a static wave impedance value or as a time-varying choice updated according to the instantaneous conditions encountered. A Lyapunov-like analysis is presented demonstrating that time variation in wave impedance will not violate the passivity of the system. Experimental trials, both in simulation and on a haptic feedback device, are presented validating the technique. Consideration is also given to the case of an uncertain environment, in which an a priori impedance choice may not be possible.