818 research outputs found
Seabed foraging by Antarctic krill: Implications for stock assessment, bentho-pelagic coupling, and the vertical transfer of iron
No full textA compilation of more than 30 studies shows that adult Antarctic krill (Euphausia superba) may frequent benthic habitats year-round, in shelf as well as oceanic waters and throughout their circumpolar range. Net and acoustic data from the Scotia Sea show that in summer 2-20% of the population reside at depths between 200 and 2000 m, and that large aggregations can form above the seabed. Local differences in the vertical distribution of krill indicate that reduced feeding success in surface waters, either due to predator encounter or food shortage, might initiate such deep migrations and results in benthic feeding. Fatty acid and microscopic analyses of stomach content confirm two different foraging habitats for Antarctic krill: the upper ocean, where fresh phytoplankton is the main food source, and deeper water or the seabed, where detritus and copepods are consumed. Krill caught in upper waters retain signals of benthic feeding, suggesting frequent and dynamic exchange between surface and seabed. Krill contained up to 260 nmol iron per stomach when returning from seabed feeding. About 5% of this iron is labile, i.e., potentially available to phytoplankton. Due to their large biomass, frequent benthic feeding, and acidic digestion of particulate iron, krill might facilitate an input of new iron to Southern Ocean surface waters. Deep migrations and foraging at the seabed are significant parts of krill ecology, and the vertical fluxes involved in this behavior are important for the coupling of benthic and pelagic food webs and their elemental repositories
Strategia nierozstrzygalności w wierszu jest taki pomnik Tadeusza Różewicza
No full textThe article deals with Tadeusz Rożewicz’s poem jest taki pomnik and puts forward a new interpretation concerning the mechanisms he uses to construct meanings. The author focuses on three issues: (1) the role of photography as an interpreter, (2) intertextual encryption of key contexts for the spiritual idea of the poem, (3) the allopathic function of laughter. She also indicates components of the strategy of undecidability, as a result of which a work containing what might seem an unambiguous ideological statement launches mechanisms guarding against one-way reading.The article deals with Tadeusz Rożewicz’s poem jest taki pomnik and puts forward a new interpretation concerning the mechanisms he uses to construct meanings. The author focuses on three issues: (1) the role of photography as an interpreter, (2) intertextual encryption of key contexts for the spiritual idea of the poem, (3) the allopathic function of laughter. She also indicates components of the strategy of undecidability, as a result of which a work containing what might seem an unambiguous ideological statement launches mechanisms guarding against one-way reading
Dr. Hamid Taki Speaking at the 3rd SCLC Public Policy Symposium, March 1981
No full textDr. Hamid Taki and Jim Zogby speak during a panel discussion that was part of the Southern Christian Leadership Conference's third Public Policy Symposium. Written on verso: Dr. Hamid Taki, professor of political science at Morehouse College (right), discusses the determined fight for liberation on the part of black South Africans during SCLC's third public policy symposium. Jim Zogby, (left), director of the Palestinian Human Rights Committee, shares international panel discussion with Dr. Taki.The Atlanta University Center Robert W. Woodruff Library acknowledges the generous support of the Joseph & Evelyn Lowery Institute for Justice and Human Rights, the Joseph Echols Lowery Irrevocable Trust, and other donors in supporting the processing and digitization of Morehouse College's Joseph Echols and Evelyn Gibson Lowery Collection
Strong Scalability Studies for the 2-D and 3-D Poisson Equations on the Taki 2018 Cluster
No full textThe new 2018 nodes in the cluster taki in the UMBC High Performance Computing Facility contain two 18-core Intel Skylake CPUs and 384 GB of memory per node, connected by an EDR (Enhanced Data Rate) InfiniBand interconnect. Parallel performance studies for the memory-bound test problem of the Poisson equation in three spatial dimensions yield several conclusions for the operation of the CPU cluster in taki. Strong scalability studies demonstrate excellent performance when using multiple nodes due to the low latency of the high-performance interconnect and good speedup when using all cores of the multi-core CPUs. Weak scalability studies confirm that best throughput is achieved by using all cores on a shared-memory node. Both types of performance studies highlight that parallel code is needed to take full advantage of the large number of computational cores on the high-memory 2018 nodes.The hardware used in the computational studies is part of the UMBC High Performance Computing Facility (HPCF). The facility is supported by the U.S. National Science Foundation through the MRI program (grant nos. CNS–0821258, CNS–1228778, and OAC–1726023) and the SCREMS program (grant no. DMS–0821311), with additional substantial support from the University of Maryland, Baltimore County (UMBC). See hpcf.umbc.edu for more information on HPCF and the projects using its resources. Co-author Carlos Barajas was supported by UMBC as HPCF RA.http://hpcf-files.umbc.edu/research/papers/PoissonHPCF20192.pd
Deformed prepotential, quantum integrable system and Liouville field theory
No full textWe study the dual descriptions recently discovered for the Seiberg–Witten theory in the presence of surface operators. The Nekrasov partition function for a four-dimensional N=2 gauge theory with a surface operator is believed equal to the wave-function of the corresponding integrable system, or the Hitchin system, and is identified with the conformal block with a degenerate field via the AGT relation. We verify the conjecture by showing that the null state condition leads to the Schrödinger equations of the integrable systems. Furthermore, we show that the deformed prepotential emerging from the period integrals of the principal function corresponds to monodromy operation of the conformal block. We also give the instanton partition functions for the asymptotically free SU(2) gauge theories in the presence of the surface operator via the AGT relation. We find that these partition functions involve the counting of two- and four-dimensional instantons
Green tea consumption and cerebral white matter lesions in community-dwelling older adults without dementia
No full textPreKURA:1657On behalf of the Japan Prospective Studies Collaboration for Aging and Dementia (JPSC-AD) study group:
Toshiharu Ninomiya, Mao Shibata, Takanori Honda, Jun Hata, Tomoyuki Ohara, Masato Akiyama, Shigeyuki Nakaji, Koichi Murashita, Tatsuya Mikami, Kaori Sawada, Shintaro Yokoyama, Tetsuya Maeda, Naoki Ishizuka, Hiroshi Akasaka, Yasuo Terayama, Hisashi Yonezawa, Junko Takahashi, Kenjiro Ono, Moeko Noguchi-Shinohara, Kazuo Iwasa, Sohshi Yuki-Nozaki, Masahito Yamada, Masaru Mimura, Shogyoku Bun, Hidehito Niimura, Ryo Shikimoto, Hisashi Kida, Kenji Nakashima, Yasuyo Fukada, Hisanori Kowa, Toshiya Nakano, Kenji Wada, Masafumi Kishi, Tomoki Ozaki, Ayumi Tachibana, Yuta Yoshino, Jun-ichi Iga, Shu-ichi Ueno, Minoru Takebayashi, Naoto Kajitani, Yusuke Miyagawa, Tomohisa Ishikawa, Seiji Yuki, Ryuji Fukuhara, Asuka Koyama, Mamoru Hashimoto, Manabu Ikeda, Yoshihiro Kokubo, Kazuhiro Uchida, Midori Esaki, Yasuyuki Taki, Yasuko Tatewaki, Benjamin Thyreau, Hisako Yoshida, Kaori Muto, Yusuke Inoue, Izen Ri, Yukihide Momozawa, Chikashi Terao, Michiaki Kubo & Yutaka Kiyoharajournal articl
Sequencer: Deep LSTM for Image Classification
No full textIn recent computer vision research, the advent of the Vision Transformer
(ViT) has rapidly revolutionized various architectural design efforts: ViT
achieved state-of-the-art image classification performance using self-attention
found in natural language processing, and MLP-Mixer achieved competitive
performance using simple multi-layer perceptrons. In contrast, several studies
have also suggested that carefully redesigned convolutional neural networks
(CNNs) can achieve advanced performance comparable to ViT without resorting to
these new ideas. Against this background, there is growing interest in what
inductive bias is suitable for computer vision. Here we propose Sequencer, a
novel and competitive architecture alternative to ViT that provides a new
perspective on these issues. Unlike ViTs, Sequencer models long-range
dependencies using LSTMs rather than self-attention layers. We also propose a
two-dimensional version of Sequencer module, where an LSTM is decomposed into
vertical and horizontal LSTMs to enhance performance. Despite its simplicity,
several experiments demonstrate that Sequencer performs impressively well:
Sequencer2D-L, with 54M parameters, realizes 84.6% top-1 accuracy on only
ImageNet-1K. Not only that, we show that it has good transferability and the
robust resolution adaptability on double resolution-band.Comment: Accepted in NeurIPS 2022; camera ready editio
- …
