Search
Skip to Search Results- 2Abdi Oskouie, Mina
- 2Birkbeck, Neil Aylon Charles
- 2Cai, Zhipeng
- 2Chen, Jiyang
- 2Chowdhury, Md Solimul
- 2Chubak, Pirooz
- 74Machine Learning
- 70Reinforcement Learning
- 41Artificial Intelligence
- 36Machine learning
- 22Natural Language Processing
- 22Reinforcement learning
-
Two Irons in the Fire: Synthesizing Libraries of Programs by Optimizing an Auxiliary Function while Solving Problems
DownloadFall 2023
Program synthesis faces a significant challenge in exploring a vast program space to find a program that satisfies the user's intent. Prior studies have proposed using different methods to guide the synthesis process to address this challenge. We propose a method that offers search guidance which...
-
Fall 2019
Policy evaluation, learning value functions, is an integral part of the reinforcement learning problem. In this thesis, I propose a neural network architecture, the Two-Timescale Network (TTN), for value function approximation which utilizes linear function approximation for the value function...
-
Fall 2014
This thesis is concerned with the ultrasonic heart image segmentation problem using parametric active contour model. Most of the existing parametric models consider only either the edge or the regional information. In this thesis, we propose a new parametric active contour model considering both...
-
Uncalibrated Vision-Based Control and Motion Planning of Robotic Arms in Unstructured Environments
DownloadFall 2012
Many robotic systems are required to operate in unstructured environments. This imposes significant challenges on algorithm design. Particularly, motion control and planning algorithms should be robust to noise and outliers, because uncertainties are inevitable. In addition, independence from...
-
Fall 2022
Some real-world deployments of deep reinforcement learning (RL) may require a human-in-the-loop. Whether to ask-for-help, obtain new demonstrations and data, or handle out-of-distribution states, many methods rely on uncertainty estimates from a neural network to determine when to solicit a...
-
Fall 2019
In this thesis, we study merge-and-shrink (M&S), a flexible abstraction technique for generating heuristics for cost optimal planning. We first propose three novel merging strategies for M&S, namely, Undirected Min-Cut (UMC), Maximum Intermediate Abstraction Size Minimizing (MIASM), and Dynamic...
-
Fall 2020
This thesis is offered as a step forward in our understanding of forgetting in artificial neural networks. ANNs are a learning system loosely based on our understanding of the brain and are responsible for recent breakthroughs in artificial intelligence. However, they have been reported to be...