PhD Thesis Defense
Nathan D. Ratliff
Carnegie Mellon University

Learning to Search: Structured Prediction Techniques for Imitation Learning

Event Location: Newell Simon Hall 1507Abstract: Modern robots successfully manipulate objects, navigate rugged terrain, drive in urban settings, and play world-class chess. Unfortunately, programming these robots is challenging, time-consuming and expensive; the parameters governing their behavior are often unintuitive, even when the desired behavior is clear and easily demonstrated. Inspired by successful end-to-end learning systems [...]

PhD Thesis Proposal
Boris Sofman
Carnegie Mellon University

Online Learning Techniques for Improving Robot Navigation in Unfamiliar Domains

Event Location: Newell Simon Hall 1305Abstract: Many mobile robot applications require robots to act safely and intelligently in complex unfamiliar environments with little structure and limited or unavailable human supervision. As a robot is forced to operate in an environment that it was not engineered or trained for, various aspects of its performance will inevitably [...]

PhD Thesis Proposal
Anton Chechetka
Carnegie Mellon University

Algorithms for Answering Queries with Graphical Models

Event Location: Newell Simon Hall 1305Abstract: In numerous real world applications, from sensor networks to patient monitoring to intelligent buildings, probabilistic inference is necessary to make conclusions about the system in question in the face of uncertainty. The key problem in all those settings is to compute the probability distribution over some random variables of [...]

PhD Thesis Proposal
Ling Xu
Carnegie Mellon University

Graph Planning for Effective Environmental Coverage

Event Location: Newell Simon Hall 1305Abstract: Tasks such as street mapping and security surveillance seek a route that traverses a given space to perform a function. These task functions may involve mapping the space for accurate modeling, sensing the space for unusual activity, or processing the space for object detection. When these tasks are performed [...]

PhD Thesis Proposal
Gregory John Barlow
Carnegie Mellon University

Generalized Density-Estimate Memory for Dynamic Problems

Event Location: Newell Simon Hall 1305Abstract: Optimization systems traditionally focus on static problems, also known as offline or a priori optimization problems. Many real-world problems may be better modeled as dynamic optimization problems, also known as stochastic, in situ, or online optimization problems. In these types of problems, the fitness landscape of the search space [...]

PhD Thesis Proposal
David Silver
Carnegie Mellon University

Learning Preference Models for Complex Mobile Robotic Systems

Event Location: Newell Simon Hall 1305Abstract: Achieving robust and reliable operation even in complex unstructured environments is a central goal of field robotics. As the environments and scenarios to which robots are applied have continued to grow in complexity, so has the challenge of properly defining preferences between various actions, and the terrains they result [...]

PhD Thesis Proposal
Amir Degani
Carnegie Mellon University

A Minimalist Dynamic Climbing Robot: Modeling, Analysis and Experiments

Event Location: Newell Simon Hall 1305Abstract: Dynamics in locomotion is highly useful, as can be seen in animals and gradually in robots. For instance, chimpanzees are dynamic climbers that can reach virtually any part of a tree and even move to neighboring trees, while sloths are quasistatic climbers confined only to a few branches. Although [...]

PhD Thesis Proposal
Jean-Francois Lalonde
Carnegie Mellon University

Understanding and Recreating Visual Appearance under Natural Illumination

Event Location: Smith Hall 100Abstract: The appearance of an outdoor scene is determined to a great extent by the prevailing illumination conditions. However, most practical computer vision applications treat illumination more as a nuisance rather than a source of signal. In this thesis proposal, we suggest that we should instead embrace illumination, even in the [...]

PhD Thesis Defense
Marius Leordeanu
Carnegie Mellon University

Spectral Graph Matching, Learning, and Inference for Computer Vision

Event Location: Newell Simon Hall 1305Abstract: Several important applications in computer vision, such as 2D and 3D object matching, object category and action recognition, object category discovery, and texture discovery and analysis, require the ability to match features efficiently in the presence of background clutter and occlusion. In order to improve matching robustness and accuracy [...]