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.

Effectiveness of Milorganite® as a Repellent to Protect Ornamental and Agronomic Plants From Deer Over-Browsing

When deer populations become locally overabundant, browsing of ornamental and agronomic plants negatively affects plant establishment, survival, and productivity. Milorganite® is a slow-release, organic fertilizer produced from human sewage. We tested Milorganite® as a deer repellent on chrysanthemums (Chrysanthemums morifolium) in an urban/suburban environment, and soybeans (Gycine max) in a rural agriculture environment. Six beds of chrysanthemums at two sites were monitored for 28 to 35 days. Treatment plants received a top dressing of 104 grams of Milorganite® (1120.9 kg/ha). Milorganite® treated plants had more (P < 0.001) terminal buds and achieved greater height (P < 0.002) compared to controls at one site, however damage observed was similar at the second site. In a second experiment, 0.2-ha plots of soybeans (Glycine max) were planted on five rural properties in northeastern Georgia and monitored for ≥ 30 days. Treated areas received 269 kg/ha of Milorganite®. In 4 of 5 sites, Milorganite® delayed browsing on treated plants from 1 week to > 5 weeks post-planting. Duration of the protection appeared to be related to the difference in deer density throughout most of the study areas. Results of this study indicate Milorganite® has potential use as a deer repellent.

Wildlife Master Volunteers: A Multi-County Approach to Resolving Human-Wildlife Conflicts

The Wildlife Master (WM) Program in Colorado was modeled after the highly successful Master Gardener volunteer program. In 10 highly populated suburban counties with large rural areas surrounding the Denver Metro Area, Colorado State University (CSU) Cooperative Extension Natural Resources agents train, supervise and manage these volunteers in the identification, referral, and resolution of wildlife damage issues. High quality, research-based training is provided by university faculty and other professionals in public health, animal damage control, wildlife management and animal behavior. Inquiries are responded to mainly via telephone. Calls by concerned residents are forwarded to WMs who provide general information about human-wildlife conflicts and possible ways to resolve complaints. Each volunteer serves a minimum of 14 days on phone duty annually, calling in from a remote location to a voice mail system from which phone messages can be conveniently retrieved. Response time per call is generally less than 24 hours. During 2004, more than 2,000 phone calls, e-mail messages and walk-in requests for assistance were fielded by 100 cooperative extension WMs. Calls fielded by volunteers in one county increased five-fold during the past five years, from 100 calls to over 500 calls annually. Valued at the rate of approximately $18.00 per volunteer hour, the leveraged value of each WM was about $450 in 2005, based on 25 hours of service and training. The estimated value of the program to Colorado in 2004 was over $45,000 of in-kind service, or about one full-time equivalent faculty member. This paper describes components of Colorado’s WM Program, with guides to the set-up of similar programs in other states.

Characterization of Glycation Sites on Human Serum Albumin using Mass Spectrometry

The modification of proteins by reducing sugars is a process that occurs naturally in the body. This process, which is known as glycation, has been linked to many of the chronic complications encountered during diabetes. Glycation has also been linked to changes in the binding of human serum albumin (HSA) to several drugs and small solutes in the body. While these effects are known, there is little information that explains why these changes in binding occur. The goal of this project was to obtain qualitative and quantitative information about glycation that occurs on HSA. The first section of this dissertation examined methods that could be used to quantify and identify glycation that occurs on HSA. The extent of glycation that occurred on HSA was quantified using oxygen-18 labeling mass spectrometry and the glycation sites were identified by observing the mass-to-charge (m/z) shifts that occurred in glycated HSA. This initial investigation revealed that oxygen-18 labeling based quantitation can be improved over previous methods if a relative comparison is done with oxygen-18 labeled peptides in a control HSA sample. Similarly, the process of making m/z shift-based assignments could be improved if only the peptides that were unique to the glycated HSA samples were used with internal calibration. These techniques were used in subsequent chapters for the assignment of early and late-stage glycation products on HSA. The regions of HSA that contained the highest amount of modification were identified, quantified, and ranked in order of their relative abundance. Of the commonly reported glycation sites, the N-terminus was found to have the highest extent of modification, followed by lysines 525, 199, and 439. The relative amount of modification on lysine 281, with respect to the aforementioned residues, varied with different degrees of glycation. The oxygen-18 labeling approach used for this analysis was novel because it allowed for the simultaneous quantification of all glycation-related modifications that were occurring on HSA. As such, several arginine residues were also found to have high amounts of modification on glycated HSA.