http://hdl.handle.net/1957/8305 2013-05-18T19:43:23Z 2013-05-18T19:43:23Z Oregon State University. Dept. of Computer Science Restificar, Angelo C. Dietterich, Thomas Glen http://hdl.handle.net/1957/36398 2013-01-29T17:31:46Z 2013-01-29T00:00:00Z

Exploiting monotonicity via logistic regression in Bayesian network learning Oregon State University. Dept. of Computer Science; Restificar, Angelo C.; Dietterich, Thomas Glen An important challenge in machine learning is to find ways of learning quickly from very small amounts of training data. The only way to learn from small data samples is to constrain the learning process by exploiting background knowledge. In this report, we present a theoretical analysis on the use of constrained logistic regression for estimating conditional probability distribution in Bayesian Networks (BN) by using background knowledge in the form of qualitative monotonicity statements. Such background knowledge is treated as a set of constraints on the parameters of a logistic function during training. Our goal of finding the appropriate BN model is two-fold: (a) we want to exploit any monotonic relationship between random variables that may generally exist as domain knowledge and (b) we want to be able to address the problem of estimating the conditional distribution of a random variable with a large number of parents. We discuss variants of the logistic regression model and present an analysis on the corresponding constraints required to implement monotonicity. More importantly, we outline the problem in some of these variants in terms of the number of parameters and constraints which, in some cases, can grow exponentially with the number of parent variables. To address this problem, we present two variants of the constrained logistic regression model, M[superscipt 2b][subscript CLR] and M³[subscript CLR], in which the number of constraints required to implement monotonicity does not grow exponentially with the number of parents hence providing a practicable method for estimating conditional probabilities with very sparse data.

2013-01-29T00:00:00Z Oregon State University. Dept. of Computer Science Piorkowski, David Fleming, Scott D. Kwan, Irwin Burnett, Margaret, 1949- Scaffidi, Chris Bellamy, Rachel Jordhal, Joshua http://hdl.handle.net/1957/36082 2013-01-11T19:31:15Z 2013-01-11T00:00:00Z

The whats and hows of programmers' foraging diets Oregon State University. Dept. of Computer Science; Piorkowski, David; Fleming, Scott D.; Kwan, Irwin; Burnett, Margaret, 1949-; Scaffidi, Chris; Bellamy, Rachel; Jordhal, Joshua One of the least studied areas of Information Foraging Theory is diet: the information foragers choose to seek. For example, do foragers choose solely based on cost, or do they stubbornly pursue certain diets regardless of cost? Do their debugging strategies vary with their diets? To investigate "what" and "how" questions like these for the domain of software debugging, we qualitatively analyzed 9 professional developers' foraging goals, goal patterns, and strategies. Participants spent 50% of their time foraging. Of their foraging, 58% fell into distinct dietary patterns—mostly in patterns not previously discussed in the literature. In general, programmers' foraging strategies leaned more heavily toward enrichment than we expected, but different strategies aligned with different goal types. These and our other findings help fill the gap as to what programmers' dietary goals are and how their strategies relate to those goals. © ACM, 2013. This is the author's version of the work. It is posted here by permission of ACM for your personal use. Not for redistribution. The definitive version was published in Proc. CHI, ACM (2013); This tech report is an extended version of publication [17]: It adds Appendix A to the end.

2013-01-11T00:00:00Z Oregon State University. Dept. of Computer Science Seifert, Ethan Stumpf, Simone Herlocker, Jonathan Lee Wynn, Eleanor H. http://hdl.handle.net/1957/35864 2012-12-26T23:36:43Z 2008-01-01T00:00:00Z

Browsing for information on the web and in the file system Oregon State University. Dept. of Computer Science; Seifert, Ethan; Stumpf, Simone; Herlocker, Jonathan Lee; Wynn, Eleanor H. Browsing is one of the methods used for finding and refinding information on the web or in the file local system and there are opportunities to avoid this, particularly if that information is revisited frequently. We present empirical results from a field study contrasting patterns of browsing to local and web information and we qualify the cost that this navigation method incurs. In addition, we provide an improved method for defining revisit behavior and report on the level of revisits during our study. Our findings have implications for solution development that reduce user effort for finding and refinding information.

2008-01-01T00:00:00Z Oregon State University. Dept. of Computer Science Dietterich, Thomas Glen http://hdl.handle.net/1957/35862 2012-12-26T23:03:18Z 2006-04-30T00:00:00Z

Proposed metrics for transfer learning Oregon State University. Dept. of Computer Science; Dietterich, Thomas Glen Summary: Four proposed metrics: [1] average relative reduction in training time (sample size, number of training experiences) [2] jumpstart (initial advantage of transfer algorithm) [3] handicap (how long it takes the no-transfer algorithm to overcome the jumpstart) [4] asymptotic advantage (how much better the transfer learning algorithm does in the limit of large sample sizes) Version 4

2006-04-30T00:00:00Z