Balancing Exploration and Exploitation: A New Algorithm for Active Machine Learning

Active machine learning algorithms are used when large numbers of unlabeled examples are available and getting labels for them is costly (e.g. requiring consulting a human expert). Many conventional active learning algorithms focus on refining the decision boundary, at the expense of exploring new regions that the current hypothesis misclassifies. We propose a new active learning algorithm that balances such exploration with refining of the decision boundary by dynamically adjusting the probability to explore at each step. Our experimental results demonstrate improved performance on data sets that require extensive exploration while remaining competitive on data sets that do not. Our algorithm also shows significant tolerance of noise.

Lincoln Downtown Rotary Club

Good afternoon, and thank you for inviting me to be here with you today. One of the things I'm enjoying most in my new position as University of Nebraska Vice President of Agriculture and Natural Resources and Harlan Vice Chancellor of the Institute of Agriculture and Natural Resources is meeting the people who live in this state, in urban and rural areas, from Scottsbluff to Omaha, from Lincoln to Curtis, and at any number of towns - north, south, east, west - in between.

Plattsmouth Rotary February Meeting (Farmer's Day Program)

Good afternoon. It's a pleasure to be with you here in Plattsmouth today, and I am particularly pleased to be part of your Farmer's Day program. Because I am so new to Nebraska and the University of Nebraska Institute of Agriculture and Natural Resources, I am trying to learn as much as I can about how the Institute connects with the state, annd how you feel we can be an even better partner with Nebraska. I'm curious about our work in each Nebraska county, and because I was coming to Cass County today, I asked those I work with for some figures on how the Institute is part of the lives of Cass Countians.

Study of Micro Rotary Ultrasonic Machining

Product miniaturization for applications in fields such as biotechnology, medical devices, aerospace, optics and communications has made the advancement of micromachining techniques essential. Machining of hard and brittle materials such as ceramics, glass and silicon is a formidable task. Rotary ultrasonic machining (RUM) is capable of machining these materials. RUM is a hybrid machining process which combines the mechanism of material removal of conventional grinding and ultrasonic machining. Downscaling of RUM for micro scale machining is essential to generate miniature features or parts from hard and brittle materials. The goal of this thesis is to conduct a feasibility study and to develop a knowledge base for micro rotary ultrasonic machining (MRUM). Positive outcome of the feasibility study led to a comprehensive investigation on the effect of process parameters. The effect of spindle speed, grit size, vibration amplitude, tool geometry, static load and coolant on the material removal rate (MRR) of MRUM was studied. In general, MRR was found to increase with increase in spindle speed, vibration amplitude and static load. MRR was also noted to depend upon the abrasive grit size and tool geometry. The behavior of the cutting forces was modeled using time series analysis. Being a vibration assisted machining process, heat generation in MRUM is low which is essential for bone machining. Capability of MRUM process for machining bone tissue was investigated. Finally, to estimate the MRR a predictive model was proposed. The experimental and the theoretical results exhibited a matching trend.

SimSight: A Virtual Machine Based Dynamic Call Graph Generator

One problem with using component-based software development approach is that once software modules are reused over generations of products, they form legacy structures that can be challenging to understand, making validating these systems difficult. Therefore, tools and methodologies that enable engineers to see interactions of these software modules will enhance their ability to make these software systems more dependable. To address this need, we propose SimSight, a framework to capture dynamic call graphs in Simics, a widely adopted commercial full-system simulator. Simics is a software system that simulates complete computer systems. Thus, it performs nearly identical tasks to a real system but at a much lower speed while providing greater execution observability. We have implemented SimSight to generate dynamic call graphs of statically and dynamically linked functions in x86/Linux environment. A case study illustrates how we can use SimSight to identify sources of software errors. We then evaluate its performance using 12 integer programs from SPEC CPU2006 benchmark suite.