Narratives are central to communication and the human experience. For a computer system to understand a narrative, it must be able to identify the key facts or plot elements that describe what happened or how the world has changed. These element are called events;establishing a document’s events and the relationships...
There is a software gap in parallel processing. The short lifespan and small installation base of parallel architectures have made it economically infeasible to develop platform-specific parallel programming environments that deliver performance and programmability. One obvious solution is to build architecture-independent programming environments. But the architecture independence usually comes at...
Existing graphics systems are too large for students to study in an introductory computer graphics course. We have implemented a lightweight, object-oriented graphics system called OGS for instruction. OGS is written in Java. It demonstrates how a graphics system is implemented from scratch and is intended to help students understand...
In real-time control systems, the value of a control decision depends not
only on the correctness of the decision but also on the time when that decision
is available. Recent work in real-time decision making has used machine learning
techniques to automatically construct reactive controllers, that is, controllers
with little...
The Symbolic Probabilistic Inference (SPI) algorithm was developed by Bruce D' Ambrosio for efficient calculation of prior probabilities in belief nets [2]. Although the complexity of the SPI algorithm compares favorably with other approaches to probabilistic inference [5], its actual running time is still prohibitively long even for moderately sized...
Easy-first, a search-based structured prediction approach, has been applied to many NLP tasks including dependency parsing and coreference resolution. This approach employs a learned greedy policy (action scoring function) to make easy decisions first, which constrains the remaining decisions and makes them easier. This thesis studies the problem of learning...
Supervised learning programs, such as decision tree learners and neural networks, often must learn Boolean functions. The concept being learned may not easily be expressed in terms of the atomic features given. Constructive induction automatically produces higher level features (combinations of the atomic features), which can improve learning performance. The...
How can an agent generalize its knowledge to new circumstances? To learn
effectively an agent acting in a sequential decision problem must make intelligent action selection choices based on its available knowledge. This dissertation focuses on Bayesian methods of representing learned knowledge and develops novel algorithms that exploit the represented...
Coral reefs around the world face numerous threats, both natural and
anthropogenic, including pollution, natural disasters, invasive species, habitat destruction,
and destructive methods of fishing. Given enough time, coral can recover from natural
disasters, but anthropogenic threats decrease corals’ ability to recover from things such as
hurricanes. It is difficult...
In this work, we study the problem of learning and improving policies for probabilistic planning problems. In the first part, we train neural network policies for probabilistic planning problems modeled as factored Markov decision problems. The objective is to train problem-specific neural networks via supervised learning to imitate the action...