Autonomous robotic agents are on their way to becoming in-home personal assistants, construction assistants, and warehouse workers. The degree of autonomy of such systems is reflected by the manner in which we specify goals to them; the abstraction of low-level commands to high-level goals goes hand-in-hand with increased autonomy. In...
In an increasingly computation-driven world, algorithms and mathematical models significantly impact decision making across various fields. To foster trust and understanding, it is crucial to provide users with clear and concise explanations of the reasoning behind the results produced by computational tools, especially when recommendations appear counterintuitive. Legal frameworks in...
This dissertation delves into understanding, characterizing, and addressing dataset shift in deep learning, a pervasive issue for deployed machine learning systems. Integral aspects of the problem are examined: We start with the use of counterfactual explanations in order to characterize the behavior of deep reinforcement learning agents in visual input...
The advancement of artificial intelligence (AI) has led to transformative developments across multiple sectors, fostering innovation and redefining our interactions with technology. As AI matures and becomes integrated into society, it offers numerous opportunities to address global challenges and revolutionize a wide array of human endeavors. These advances are driven...
Robotic Bipedal locomotion holds the potential for efficient, robust traversal of difficult terrain. The difficulty lies in the dynamics of locomotion which complicate control and motion planning. Bipedal locomotion dynamics are dimensionally large problems, extremely nonlinear, and operate on the limits of actuator capabilities, which limit the performance of generic...
We explore the application of deep learning to the disparate fields of natural language processing and computational biology. Both the sentences uttered by humans as well as the RNA and protein sequences found within the cells of their bodies can be considered formal languages in computer science, as sets of...
As one of the most popular data types, the point cloud is widely used in various appli- cations, including computer vision, computer graphics and robotics. The capability to directly measure 3D point clouds is invaluable in those applications as depth information could remove a lot of the segmentation ambiguities in...
Many large-scale data analysis applications involve data that can vary over both time and space. Often the primary goal of analyzing spatiotemporal data is identifying trends, movements, and sudden changes with respect to time, location, or both. This can include a variety of applications in economics (housing prices, unemployment, job...
The ability to extract uncertainties from predictions is crucial for the adoption of deep learning systems to safety-critical applications. Uncertainty estimates can be used as a failure signal, which is necessary for automating complex tasks where safety is a concern. Furthermore, current deep learning systems do not provide uncertainty estimates,...
In this thesis, I present the variational database management system, a formal framework and its implementation for representing variation in relational databases and managing variational information needs. A variational database is intended to support any kind of variation in a database. Specific kinds of variation in databases have already been...
This dissertation addresses few-shot object segmentation in images. The goal of segmentation is to label every image pixel with a class of the object occupying that pixel, where the class may represent a semantic object category or instance. In few-shot segmentation, training and test datasets have different classes. Every new...
In this dissertation, we address action segmentation in videos under limited supervision. The goal of action segmentation is to predict an action class for each frame of a video. The limited supervision means ground truth labels of video frames are not available in training. We focus on three types of...
Anomaly detection aims at detecting the points that appear different than the majority of the data, such that they are suspected to be generated from a different distribution. Anomaly detectors have been applied in many different fields, such as detecting fraudulent behaviors in bank transaction, finding broken sensors in a...
Advances in sensor technology are greatly expanding the range of quantities that can be measured while simultaneously reducing the cost. However, deployed sensors drift out of calibration and fail, so every sensor network requires quality control procedures to promptly detect these failures. To address these problems, we propose a two-level...
Humans are remarkably efficient in learning by interacting with other people and observing their behavior. Children learn by watching their parents’ actions and mimic their behavior. When they are not sure about their parents demonstration, they communicate with them, ask questions, and learn from their feedback. On the other hand,...
Scientists and engineers have to analyze and query multiple large databases. Analysis over databases created by phasor measurement units can provide insight into the health of the grid, thereby improving control over operations. Realizing this data-driven control, however, requires validating, processing and storing massive amounts of PMU data efficiently, which...
Learning novel concepts from relational databases is an important problem with applications in several disciplines, such as data management, natural language processing, and bioinformatics. For a learning algorithm to be effective, the input data should be clean and in some desired representation. However, real-world data is usually heterogeneous – the...
There are nearly two million limb amputees living in the United States of America. Loss of limbs results in profound changes in one's life. However, the underlying neural circuitry and much of the ability to sense and control movements of their missing limb is retained even after limb loss. This...
This dissertation addresses object recognition in challenging settings, where distinct object classes are visually very similar (e.g., species of birds and insects) and/or access to training examples of object classes is limited (e.g., due to the associated high costs of data annotation). In this dissertation, we present a variety of...
The thesis focuses on activity recognition from sensor data, which has spurred a great deal of interest due to its impact on health care and security. Previous work on activity recognition from multivariate time series data has mainly applied supervised learning techniques which require a high degree of annotation effort...
Most tasks in natural language processing (NLP) try to map structured input (e.g., sentence or word sequence) to some form of structured output (tag sequence, parse tree, semantic graph, translated/paraphrased/compressed sentence), a problem known as “structured prediction”. While various learning algorithms such as the perceptron, maximum entropy, and expectation-maximization have...
Markov Decision Processes (MDPs) are the de-facto formalism for studying sequential decision making problems with uncertainty, ranging from classical problems such as inventory control and path planning, to more complex problems such as reservoir control under rainfall uncertainty and emergency response optimization for fire and medical emergencies. Most prior research...
Machine learning models for natural language processing have traditionally relied on large numbers of discrete features, built up from atomic categories such as word forms and part-of-speech labels, which are considered completely distinct from each other. Recently however, the advent of dense feature representations coupled with deep learning techniques has...
Recognizing human actions in videos is a long-standing problem in computer vision with a wide range of applications including video surveillance, content retrieval, and sports analysis. This thesis focuses on addressing efficiency and robustness of video classification in unconstrained real-world settings. The thesis work can be broadly divided into four...
Software testing is a very important task during software development and it can be used to improve the quality and reliability of the software system. One potential way to reduce the cost and increase the efficiency of software testing is to generate test data automatically. Search-based approaches successfully generate unit...
In this work, we study network coding technique, its relation to random matrices, and their applications to communication systems. The dissertation consists of three main contributions. First, we propose efficient algorithms for data synchronization via a broadcast channel using random network coding. Second, we study the resiliency of network coding...
Machine learning systems are generally trained offline using ground truth data that has been labeled by experts. However, these batch training methods are not a good fit for many applications, especially in the cases where complete ground truth data is not available for offline training. In addition, batch methods do...
In real networks, identifying dense regions is of great importance. For example, in a network that represents academic collaboration, authors within the densest component of the graph tend to be the most prolific. Dense subgraphs often identify communities in social networks. And dense subgraphs can be used to discover regulatory...
This dissertation addresses the problem of recognizing human activities in videos. Our focus is on activities with stochastic structure, where the activities are characterized by variable space-time arrangements of actions, and conducted by a variable number of actors. These activities occur frequently in sports and surveillance videos. They may appear...
Tensegrity structures are composed of pure compressional elements that are connected via a network of pure tensional elements. The concept of tensegrity promises numerous advantages to the field of robotics. Tensegrity robots are, however, notoriously difficult to control due to their oscillatory nature and nonlinear interaction between the components. Multiagent...
Citizen Science is a paradigm in which volunteers from the general public participate in scientific studies, often by performing data collection. This paradigm is especially useful if the scope of the study is too broad to be performed by a limited number of trained scientists. Although citizen scientists can contribute...
Partial programming is a field of study where users specify an outline or skeleton of a program, but leave various parts undefined. The undefined parts are then completed by an external mechanism to form a complete program. Adaptation-Based Programming (ABP) is a method of partial programming that utilizes techniques from...
Bayesian Optimization (BO) methods are often used to optimize an unknown function f(•) that is costly to evaluate. They typically work in an iterative manner. In each iteration, given a set of observation points, BO algorithms select k ≥ 1 points to be evaluated. The results of those points are...
Networks of distributed, remote sensors are providing ecological scientists with a view of our environment that is unprecedented in detail. However, these networks are subject to harsh conditions, which lead to malfunctions in individual sensors and failures in network communications. This behavior manifests as corrupt or missing measurements in the...
Acting intelligently to efficiently solve sequential decision problems requires the ability to extract hierarchical structure from the underlying domain dynamics, exploit it for optimal or near-optimal decision-making, and transfer it to related problems instead of solving every problem in isolation. This dissertation makes three contributions toward this goal.
The first...
This dissertation addresses a number of inter-related and fundamental problems in computer vision. Specifically, we address object discovery, recognition, segmentation, and 3D pose estimation in images, as well as 3D scene reconstruction and scene interpretation. The key ideas behind our approaches include using shape as a basic object feature, and...
This dissertation addresses two fundamental problems in computer vision—namely,
multitarget tracking and event recognition in videos. These problems are challenging
because uncertainty may arise from a host of sources, including motion blur,
occlusions, and dynamic cluttered backgrounds. We show that these challenges can be
successfully addressed by using a multiscale,...
We investigate a number of techniques for increasing throughput and quality of media applications over wireless networks. A typical media communication application such as video streaming imposes strict requirements on the delay and throughout of its packets, which unfortunately, cannot be guaranteed by the underlying wireless network due inherently to...
Reinforcement learning in real-world domains suffers from three curses of dimensionality: explosions in state and action spaces, and high
stochasticity or "outcome space" explosion. Multiagent domains are particularly susceptible to these problems. This thesis describes ways to mitigate these curses in several different multiagent domains, including real-time delivery of products...
The use of autonomous robots in complex exploration tasks is rapidly increasing. Indeed, robots can provide speed and cost effectiveness in many tasks, as well as allow operation in environments that are hostile to humans. In this dissertation we: 1) provide two adaptive navigation algorithms; 2) develop a coordination mechanism;...
This dissertation explores the idea of applying machine learning technologies to help computer users find information and better organize electronic resources, by presenting the research work conducted in the following three applications: FolderPredictor, Stacking Recommendation Engines, and Integrating Learning and Reasoning.
FolderPredictor is an intelligent desktop software tool that helps...
Sequential supervised learning problems arise in many real applications. This dissertation focuses on two important research directions in sequential supervised learning: efficient training and feature induction.
In the direction of efficient training, we study the training of conditional random fields (CRFs), which provide a flexible and powerful model for sequential...
Until a few years ago, wireless-capable laptops were considered novelties by many. It is now hard to find a laptop or a hand-held computing device that is not wireless-ready. As wireless devices are becoming commodities, they have also become an indispensable part of the modern society. Not surprisingly, research in...
Knowledge workers are struggling in the information flood. There is a growing interest in intelligent desktop environments that help knowledge workers organize their daily life. Intelligent desktop environments allow the desktop user to define a set of “activities” that characterize the user’s desktop work. These environments then attempt to identify...
Automated recognition of object categories in images is a critical step for many real-world computer vision applications. Interest region detectors and region descriptors have been widely employed to tackle the variability of objects in pose, scale, lighting, texture, color, and so on. Different types of object recognition problems usually require...
In this dissertation, we present a user-in-the-loop method for the design of an interactive motion data structure that benefits from the advantages of both motion graphs and blend-based techniques. Our novel approach automatically analyzes a traditional motion graph built from labeled motion clips. The result is a more condensed, coarser...
For a certain class of Z²-actions, we provide a proof of a conjecture that the ratio of the Perron eigenvalues of the transfer matrices of the free boundary restrictions converge to the entropy of that action. Also, a novel method for computing the entropy of Z²-actions is conjectured.
Building intelligent computer assistants has been a long-cherished goal of AI. Many intelligent assistant systems were built and fine-tuned to specific application domains. In this work, we develop a general model of assistance that combines three powerful ideas: decision theory, hierarchical task models and probabilistic relational languages. We use the...