University of California, Santa Barbara

About: University of California, Santa Barbara is a education organization based out in Santa Barbara, California, United States. It is known for research contribution in the topics: Population & Galaxy. The organization has 30281 authors who have published 80852 publications receiving 4626827 citations. The organization is also known as: UC Santa Barbara & UCSB.

Deterministic annealing for clustering, compression, classification, regression, and related optimization problems

Abstract: The deterministic annealing approach to clustering and its extensions has demonstrated substantial performance improvement over standard supervised and unsupervised learning methods in a variety of important applications including compression, estimation, pattern recognition and classification, and statistical regression. The application-specific cost is minimized subject to a constraint on the randomness of the solution, which is gradually lowered. We emphasize the intuition gained from analogy to statistical physics. Alternatively the method is derived within rate-distortion theory, where the annealing process is equivalent to computation of Shannon's rate-distortion function, and the annealing temperature is inversely proportional to the slope of the curve. The basic algorithm is extended by incorporating structural constraints to allow optimization of numerous popular structures including vector quantizers, decision trees, multilayer perceptrons, radial basis functions, and mixtures of experts.

Effects of drying–rewetting frequency on soil carbon and nitrogen transformations

Abstract: Soil drying and rewetting impose a significant stress on the soil microbial community. While wetting events are common in most environments, the short and long-term effects of soil rewetting on microbial processes have not been well studied. Furthermore, it is not clear if stress history is important to consider when modeling microbial controls on ecosystem dynamics. In this experiment, we manipulated the frequency of soil rewetting events during 2 months to determine how stress history influences the response of soil microbial communities to rewetting events. Two soils were collected from the Sedgwick Ranch Natural Reserve in Santa Ynez, CA, one from an annual grassland, the other from underneath an oak canopy. Soils were incubated in the lab and went through either 0, 1, 2, 4, 6, 9, or 15 drying–rewetting cycles over 2 months. Soil moisture content was adjusted so that the average moisture content over the course of the incubation was the same for all samples, compensating for the number of drying–rewetting cycles. Soils were analyzed for respiration rate, substrate utilization efficiency, nitrification potential, microbial biomass, and NH4+ and NO3− concentrations. Total CO2 loss during incubation significantly increased with number of rewetting events for oak soils but not for grass soils, where a large number of rewetting events decreased total CO2 loss. Exposure to frequent drying–rewetting events decreased the amount of CO2 released upon rewetting and dramatically increased the activity of autotrophic nitrifier populations. For up to 6 weeks after the last drying–rewetting cycle, respiration rates in soils exposed to a history of drying–rewetting events were substantially lower than their non-stressed controls. In all cases, the effects of the rewetting stress were greater in oak than in grass soils. The results indicate that drying–rewetting events can induce significant changes in microbial C and N dynamics and these effects can last for more than a month after the last stress. The frequency of drying–rewetting stress events has important ecosystem-level ramifications and should be incorporated into models of soil microbial dynamics.

Comparison of Riemann and Lebesgue sampling for first order stochastic systems

Abstract: The normal approach to digital control is to sample periodically in time. Using an analog of integration theory we can call this Riemann sampling. Lebesgue sampling or event based sampling is an alternative to Riemann sampling. It means that signals are sampled only when measurements pass certain limits. In this paper it is shown that Lebesgue sampling gives better performance for some simple systems.

Amyloid ion channels: A common structural link for protein-misfolding disease

Abstract: Protein conformational diseases, including Alzheimer's, Huntington's, and Parkinson's diseases, result from protein misfolding, giving a distinct fibrillar feature termed amyloid. Recent studies show that only the globular (not fibrillar) conformation of amyloid proteins is sufficient to induce cellular pathophysiology. However, the 3D structural conformations of these globular structures, a key missing link in designing effective prevention and treatment, remain undefined as of yet. By using atomic force microscopy, circular dichroism, gel electrophoresis, and electrophysiological recordings, we show here that an array of amyloid molecules, including amyloid-β(1–40), α-synuclein, ABri, ADan, serum amyloid A, and amylin undergo supramolecular conformational change. In reconstituted membranes, they form morphologically compatible ion-channel-like structures and elicit single ion-channel currents. These ion channels would destabilize cellular ionic homeostasis and hence induce cell pathophysiology and degeneration in amyloid diseases.