Hasil untuk "q-bio.OT"

H. Srivastava, Choi Junesang

R. Koekoek, Rene F. Swarttouw

A system for automatically reading symbols, preferably figures, which are hand-written on an information carrier in an arrangement of squares provided on the information carrier. The images of these symbols are converted by an image convertor of glass fiber bundles to fit a camera tube screen where they are scanned vertically, quantized, and encoded to determine the size and numerical locations of intersections of the scanning beam with the lines in each symbol in each rectangle. This information is then processed by being stored and first roughly classified according to the maximum number of these intersections per symbol, each of which classes are then more specifically classified by being further processed as to the location of the mergings of the intersections, if any, in the upper, lower, right, and/or left part of the symbols, as well as determining the shape, length and/or width of the lines in certain of the symbols for their specific recognition, or identification. This recognized information then may be used for punching a code into the information carrier. If desired, the processor of this information can be located remote from the viewer and punching apparatus.

Junling Hu, Michael P. Wellman

A. Bharadwaj, Sundar G. Bharadwaj, B. Konsynski

M. C. Davey, D. Mackay

A. Heister, S. Schael, R. Barate et al.

Steven R. Brown

Martin A. Riedmiller

J. Sinova, J. Sinova, D. Culcer et al.

We describe a new effect in semiconductor spintronics that leads to dissipationless spin currents in paramagnetic spin-orbit coupled systems. We argue that in a high-mobility two-dimensional electron system with substantial Rashba spin-orbit coupling, a spin current that flows perpendicular to the charge current is intrinsic. In the usual case where both spin-orbit split bands are occupied, the intrinsic spin-Hall conductivity has a universal value for zero quasiparticle spectral broadening.

Q. Wan, Quanshun Li, Y. Chen et al.

Simon Watts, P. Stenner

Jingda Wu, Hongwen He, Jiankun Peng et al.

Abstract Reinforcement learning is a new research hotspot in the artificial intelligence community. Q learning as a famous reinforcement learning algorithm can achieve satisfactory control performance without need to clarify the complex internal factors in controlled objects. However, discretization state is necessary which limits the application of Q learning in energy management for hybrid electric bus (HEB). In this paper the deep Q learning (DQL) is adopted for energy management issue and the strategy is proposed and verified. Firstly, the system modeling of bus configuration are described. Then, the energy management strategy based on deep Q learning is put forward. Deep neural network is employed and well trained to approximate the action value function (Q function). Furthermore, the Q learning strategy based on the same model is mentioned and applied to compare with deep Q learning. Finally, a part of trained decision network is analyzed separately to verify the effectiveness and rationality of the DQL-based strategy. The training results indicate that DQL-based strategy makes a better performance than that of Q learning in training time consuming and convergence rate. Results also demonstrate the fuel economy of proposed strategy under the unknown driving condition achieves 89% of dynamic programming-based method. In addition, the technique can finally learn to the target state of charge under different initial conditions. The main contribution of this study is to explore a novel reinforcement learning methodology into energy management for HEB which solve the curse of state variable dimensionality, and the techniques can be adopted to solve similar problems.

W. Xie, Chunhai Chen, Zezhong Yang et al.

Abstract The sweetpotato whitefly Bemisia tabaci is a highly destructive agricultural and ornamental crop pest. It damages host plants through both phloem feeding and vectoring plant pathogens. Introductions of B. tabaci are difficult to quarantine and eradicate because of its high reproductive rates, broad host plant range, and insecticide resistance. A total of 791 Gb of raw DNA sequence from whole genome shotgun sequencing, and 13 BAC pooling libraries were generated by Illumina sequencing using different combinations of mate-pair and pair-end libraries. Assembly gave a final genome with a scaffold N50 of 437 kb, and a total length of 658 Mb. Annotation of repetitive elements and coding regions resulted in 265.0 Mb TEs (40.3%) and 20 786 protein-coding genes with putative gene family expansions, respectively. Phylogenetic analysis based on orthologs across 14 arthropod taxa suggested that MED/Q is clustered into a hemipteran clade containing A. pisum and is a sister lineage to a clade containing both R. prolixus and N. lugens. Genome completeness, as estimated using the CEGMA and Benchmarking Universal Single-Copy Orthologs pipelines, reached 96% and 79%. These MED/Q genomic resources lay a foundation for future ‘pan-genomic’ comparisons of invasive vs. noninvasive, invasive vs. invasive, and native vs. exotic Bemisia, which, in return, will open up new avenues of investigation into whitefly biology, evolution, and management.

Zinan Wang, Li Zhang, Song Wang et al.

We demonstrate a novel distributed acoustic sensing (DAS) system based on phase-sensitive optical time-domain reflectometry (Φ-OTDR). Both the phase and the amplitude of the Rayleigh scattering (RS) light can be demodulated in real-time. The technique is based on I/Q demodulation and homodyne detection using a 90° optical hybrid. The theoretical analysis is given, and as a proof of the concept, the dynamic strain sensing is experimentally demonstrated, with a sensing range of 12.566 km and a spatial resolution of 10 m.

Satwinder Singh, Naeem Tabassum, Tamer K. Darwish et al.

K. Parsons, D. Calic, M. Pattinson et al.

Qingfeng Lan, Yangchen Pan, Alona Fyshe et al.

Q-learning suffers from overestimation bias, because it approximates the maximum action value using the maximum estimated action value. Algorithms have been proposed to reduce overestimation bias, but we lack an understanding of how bias interacts with performance, and the extent to which existing algorithms mitigate bias. In this paper, we 1) highlight that the effect of overestimation bias on learning efficiency is environment-dependent; 2) propose a generalization of Q-learning, called \emph{Maxmin Q-learning}, which provides a parameter to flexibly control bias; 3) show theoretically that there exists a parameter choice for Maxmin Q-learning that leads to unbiased estimation with a lower approximation variance than Q-learning; and 4) prove the convergence of our algorithm in the tabular case, as well as convergence of several previous Q-learning variants, using a novel Generalized Q-learning framework. We empirically verify that our algorithm better controls estimation bias in toy environments, and that it achieves superior performance on several benchmark problems.

S. Banasick

Q Methodology is an approach to understanding subjectivity that combines qualitative and quantitative techniques (Brown, 1996 Ramlo:2016). Originally developed in the 1930s, it allows for a systematic investigation into the viewpoints or perspectives of the participants in the study (Watts & Stenner, 2012). A Q methodology study begins with the researcher assembling a set of statements related to the research topic. The statements are often drawn from participant interviews, but can also be derived from theories related to the research topic or other sources (Brown, 1996). The participants in the study are asked to rank and sort the statements in accordance with a predefined grid pattern (Figure 1). If the participants feel that the statement aligns with their opinion they are asked to place it more to the right (positive) side of the grid, while if they disagree with it they should place it more to the left (negative) side.

Yanwei Jia, X. Zhou

We study the continuous-time counterpart of Q-learning for reinforcement learning (RL) under the entropy-regularized, exploratory diffusion process formulation introduced by Wang et al. (2020). As the conventional (big) Q-function collapses in continuous time, we consider its first-order approximation and coin the term ``(little) q-function". This function is related to the instantaneous advantage rate function as well as the Hamiltonian. We develop a ``q-learning"theory around the q-function that is independent of time discretization. Given a stochastic policy, we jointly characterize the associated q-function and value function by martingale conditions of certain stochastic processes, in both on-policy and off-policy settings. We then apply the theory to devise different actor-critic algorithms for solving underlying RL problems, depending on whether or not the density function of the Gibbs measure generated from the q-function can be computed explicitly. One of our algorithms interprets the well-known Q-learning algorithm SARSA, and another recovers a policy gradient (PG) based continuous-time algorithm proposed in Jia and Zhou (2022b). Finally, we conduct simulation experiments to compare the performance of our algorithms with those of PG-based algorithms in Jia and Zhou (2022b) and time-discretized conventional Q-learning algorithms.

Zhengyu Song

This study aimed to elucidate the oncogenic role of NSUN6 and its downstream molecular network in breast cancer cells,with a focus on proliferation and migration for the key drivers of tumor progression.