University of Massachusetts Amherst

About: University of Massachusetts Amherst is a education organization based out in Amherst Center, Massachusetts, United States. It is known for research contribution in the topics: Population & Galaxy. The organization has 37274 authors who have published 83965 publications receiving 3834996 citations. The organization is also known as: UMass Amherst & Massachusetts State College.

Star formation and feedback in smoothed particle hydrodynamic simulations – I. Isolated galaxies

Abstract: We present an analysis of star formation and feedback recipes appropriate for galactic smoothed particle hydrodynamics simulations. Using an isolated Milky Way-like galaxy, we constrain these recipes based on well-established observational results. Our star formation recipe is based on that of Katz (1992) with the additional inclusion of physically motivated supernova feedback recipes. We propose a new feedback recipe in which type II supernovae are modelled using an analytical treatment of blastwaves. With this feedback mechanism and a tuning of other star formation parameters, the star formation in our isolated Milky Way-like galaxy follows the slope and normalisation of the observed Schmidt law. In addition, we reproduce the low density cutoff and filamentary structure of star formation observed in disk galaxies. Our final recipe will enable better comparison of N-body simulations with observations.

Linear least-squares algorithms for temporal difference learning

Abstract: We introduce two new temporal diffence (TD) algorithms based on the theory of linear least-squares function approximation. We define an algorithm we call Least-Squares TD (LS TD) for which we prove probability-one convergence when it is used with a function approximator linear in the adjustable parameters. We then define a recursive version of this algorithm, Recursive Least-Squares TD (RLS TD). Although these new TD algorithms require more computation per time-step than do Sutton's TD(λ) algorithms, they are more efficient in a statistical sense because they extract more information from training experiences. We describe a simulation experiment showing the substantial improvement in learning rate achieved by RLS TD in an example Markov prediction problem. To quantify this improvement, we introduce theTD error variance of a Markov chain, ωTD, and experimentally conclude that the convergence rate of a TD algorithm depends linearly on ωTD. In addition to converging more rapidly, LS TD and RLS TD do not have control parameters, such as a learning rate parameter, thus eliminating the possibility of achieving poor performance by an unlucky choice of parameters.

Macroscopic 10-terabit-per-square-inch arrays from block copolymers with lateral order.

Abstract: Generating laterally ordered, ultradense, macroscopic arrays of nanoscopic elements will revolutionize the microelectronic and storage industries. We used faceted surfaces of commercially available sapphire wafers to guide the self-assembly of block copolymer microdomains into oriented arrays with quasi-long-range crystalline order over arbitrarily large wafer surfaces. Ordered arrays of cylindrical microdomains 3 nanometers in diameter, with areal densities in excess of 10 terabits per square inch, were produced. The sawtoothed substrate topography provides directional guidance to the self-assembly of the block copolymer, which is tolerant of surface defects, such as dislocations. The lateral ordering and lattice orientation of the single-grain arrays of microdomains are maintained over the entire surface. The approach described is parallel, applicable to different substrates and block copolymers, and opens a versatile route toward ultrahigh-density systems.

Multiprocessor Scheduling with the Aid of Network Flow Algorithms

Abstract: In a distributed computing system a modular program must have its modules assigned among the processors so as to avoid excessive interprocessor communication while taking advantage of specific efficiencies of some processors in executing some program modules. In this paper we show that this program module assignment problem can be solved efficiently by making use of the well-known Ford–Fulkerson algorithm for finding maximum flows in commodity networks as modified by Edmonds and Karp, Dinic, and Karzanov. A solution to the two-processor problem is given, and extensions to three and n-processors are considered with partial results given without a complete efficient solution.

Polarized cell growth in higher plants.

Abstract: Pollen tubes and root hairs are highly elongated, cylindrically shaped cells whose polarized growth permits them to explore the environment for the benefit of the entire plant. Root hairs create an enormous surface area for the uptake of water and nutrients, whereas pollen tubes deliver the sperm cells to the ovule for fertilization. These cells grow exclusively at the apex and at prodigious rates (in excess of 200 nm/s for pollen tubes). Underlying this rapid growth are polarized ion gradients and fluxes, turnover of cytoskeletal elements (actin microfilaments), and exocytosis and endocytosis of membrane vesicles. Intracellular gradients of calcium and protons are spatially localized at the growing apex; inward fluxes of these ions are apically directed. These gradients and fluxes oscillate with the same frequency as the oscillations in growth rate but not with the same phase. Actin microfilaments, which together with myosin generate reverse fountain streaming, undergo rapid turnover in the apical domain, possibly being regulated by key actin-binding proteins, e.g., profilin, villin, and ADF/cofilin, in concert with the ion gradients. Exocytosis of vesicles at the apex, also dependent on the ion gradients, provides precursor material for the continuously expanding cell wall of the growing cell. Elucidation of the interactions and of the dynamics of these different components is providing unique insight into the mechanisms of polarized growth.