Most plant toxicology tests developed in support of environmental laws use a single stress applied to an individual plant. While tests using individual species or stresses require fewer resources and are easier to interpret, they are under increasing criticism for being unrealistic and missing important ecological interactions. The objective of...
A programming paradigm can be defined as a model or an approach employed in solving a problem. The results of the research described in this document demonstrate that it is possible to unite several different programming paradigms into a single linguistic framework. The imperative, procedural, applicative, lambda-free, relational, logic and...
Orchardgrass (Dactylis glomerata L.) is an important seed crop, but unlike other cool-season perennial grass seed crops such as perennial ryegrass (Lolium perenne L.) and tall fescue [Schedonorus arundinaceus (Schreb.) Dumort.], seed yields have not increased over time so there is considerable room for improvement. Research suggests that plant growth...
Seed shattering is a constraint to improvement of seed yield in Lolium perenne L. and as a result, seed retention is an essential trait for this valuable crop. Field studies were conducted near Corvallis, OR in two years (2017-2018 and 2018-2019) to identify genetic variation for seed retention and to...
One important cool-season perennial grass seed crop grown in Oregon is fine fescue (Festuca rubra L.). Open-field burning has been used for many decades to manage pests, cycle nutrients, and stabilize yield, in fine fescue seed fields, especially as stands age. Legislative restrictions currently limit open-field burning to 6,070 hectares...
Summer drought is a major factor limiting the regrowth of perennial ryegrass seed crops. This phase of crop development has a strong influence on seed yield because most of the tillers that contribute towards next season's seed crop are produced or regrown during this period. In recent years many seed...
Sequential supervised learning problems arise in many real applications. This dissertation focuses on two important research directions in sequential supervised learning: efficient training and feature induction.
In the direction of efficient training, we study the training of conditional random fields (CRFs), which provide a flexible and powerful model for sequential...
Advances in sensor technology are greatly expanding the range of quantities that can be measured while simultaneously reducing the cost. However, deployed sensors drift out of calibration and fail, so every sensor network requires quality control procedures to promptly detect these failures. To address these problems, we propose a two-level...
Machine learning systems are generally trained offline using ground truth data that has been labeled by experts. However, these batch training methods are not a good fit for many applications, especially in the cases where complete ground truth data is not available for offline training. In addition, batch methods do...