检索结果-内蒙古大学图书馆

End-to-end data-driven weather prediction

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Vaughan, Anna Markou, Stratis Tebbutt, Will Requeima, James Bruinsma, Wessel P. Andersson, Tom R. Herzog, Michael Lane, Nicholas D. Chantry, Matthew Hosking, J. Scott Turner, Richard E. Department of Computer Science and Technology University of Cambridge Cambridge United Kingdom Department of Engineering University of Cambridge Cambridge United Kingdom Vector Institute University of Toronto Toronto Canada Microsoft Research Cambridge United Kingdom British Antarctic Survey Cambridge United Kingdom Department of Geography University of Cambridge Cambridge United Kingdom The Alan Turing Institute London United Kingdom European Centre for Medium-Range Weather Forecasts Reading United Kingdom Google Deepmind London United Kingdom

Weather forecasting is critical for a range of human activities including transportation, agriculture, industry, as well as the safety of the general public. Machine learning models have the potential to transform the complex weather prediction pipeline, but current approaches still rely on numerical weather prediction (NWP) systems, limiting forecast speed and accuracy. Here we demonstrate that a machine learning model can replace the entire operational NWP pipeline. Aardvark Weather, an end-to-end data-driven weather prediction system, ingests raw observations and outputs global gridded forecasts and local station forecasts. Further, it can be optimised end-to-end to maximise performance over quantities of interest. Global forecasts outperform an operational NWP baseline for multiple variables and lead times. Local station forecasts are skillful up to ten days lead time and achieve comparable and often lower errors than a post-processed global NWP baseline and a state-of-the-art end-to-end forecasting system with input from human forecasters. These forecasts are produced with a remarkably simple neural process model using just 8% of the input data and three orders of magnitude less compute than existing NWP and hybrid AI-NWP methods. We anticipate that Aardvark Weather will be the starting point for a new generation of end-to-end machine learning models for medium-range forecasting that will reduce computational costs by orders of magnitude and enable the rapid and cheap creation of bespoke models for users in a variety of fields, including for the developing world where state-of-the-art local models are not currently available. © 2024, CC BY.

关键词： Weather forecasting

DL3DV-10K: A Large-Scale Scene Dataset for Deep Learning-based 3D Vision

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Ling, Lu Sheng, Yichen Tu, Zhi Zhao, Wentian Xin, Cheng Wan, Kun Yu, Lantao Guo, Qianyu Yu, Zixun Lu, Yawen Li, Xuanmao Sun, Xingpeng Ashok, Rohan Mukherjee, Aniruddha Kang, Hao Kong, Xiangrui Hua, Gang Zhang, Tianyi Benes, Bedrich Bera, Aniket Department of Computer Science Purdue University United States Adobe Inc. United States Rutgers University United States Google Inc. United States Huazhong University of Science and Technology China Wormpex AI Research

We have witnessed significant progress in deep learning-based 3D vision, ranging from neural radiance field (NeRF) based 3D representation learning to applications in novel view synthesis (NVS). However, existing scene-level datasets for deep learning-based 3D vision, limited to either synthetic environments or a narrow selection of real-world scenes, are quite insufficient. This insufficiency not only hinders a comprehensive benchmark of existing methods but also caps what could be explored in deep learning-based 3D analysis. To address this critical gap, we present DL3DV-10K, a large-scale scene dataset, featuring 51.2 million frames from 10,510 videos captured from 65 types of point-of-interest (POI) locations, covering both bounded and unbounded scenes, with different levels of reflection, transparency, and lighting. We conducted a comprehensive benchmark of recent NVS methods on DL3DV-10K, which revealed valuable insights for future research in NVS. In addition, we have obtained encouraging results in a pilot study to learn generalizable NeRF from DL3DV-10K, which manifests the necessity of a large-scale scene-level dataset to forge a path toward a foundation model for learning 3D representation. Our DL3DV-10K dataset, benchmark results, and models will be publicly accessible at Project Page. Copyright © 2023, The Authors. All rights reserved.

关键词： Large dataset

Learning noise via dynamical decoupling of entangled qubits

学校读者我要写书评

暂无评论

Physical Review A 2023年第5期107卷 052610-052610页

作者： Trevor McCourt Charles Neill Kenny Lee Chris Quintana Yu Chen Julian Kelly Jeffrey Marshall V. N. Smelyanskiy M. I. Dykman Alexander Korotkov Isaac L. Chuang A. G. Petukhov Department of Electrical Engineering and Computer Science Massachusetts Institute of Technology Cambridge Massachusetts 02139 USA Department of Physics Co-Design Center for Quantum Advantage Massachusetts Institute of Technology Cambridge Massachusetts 02139 USA Google Quantum AI Santa Barbara California 93101 USA QuAIL NASA Ames Research Center Moffett Field California 94035 USA USRA Research Institute for Advanced Computer Science Mountain View California 94043 USA Department of Physics and Astronomy Michigan State University East Lansing Michigan 48824 USA Department of Physics Massachusetts Institute of Technology Cambridge Massachusetts 02139 USA

Understanding noise in entangled systems is a prerequisite for developing scalable quantum computers. Here, we propose and apply multiqubit dynamical decoupling sequences that characterize noise during two-qubit gates. This noise is qualitatively different from the well-studied noise that leads to single-qubit dephasing; it simultaneously affects the two qubits, inducing fluctuations in their entangling parameter. In our superconducting system, the experimentally observed noise comes from coupler flux fluctuations and is observed to be non-Gaussian, leading to the stepwise decay of signals.

关键词： Quantum benchmarking Quantum coherence & coherence measures Quantum control Quantum engineering Quantum sensing

A Contrastive Framework for Neural Text Generation

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Su, Yixuan Lan, Tian Wang, Yan Yogatama, Dani Kong, Lingpeng Collier, Nigel Language Technology Lab University of Cambridge United Kingdom Tencent Ai Lab United States DeepMind United Kingdom Department of Computer Science The University of Hong Kong Hong Kong

Text generation is of great importance to many natural language processing applications. However, maximization-based decoding methods (e.g., beam search) of neural language models often lead to degenerate solutions-the generated text is unnatural and contains undesirable repetitions. Existing approaches introduce stochasticity via sampling or modify training objectives to decrease the probabilities of certain tokens (e.g., unlikelihood training). However, they often lead to solutions that lack coherence. In this work, we show that an underlying reason for model degeneration is the anisotropic distribution of token representations. We present a contrastive solution: (i) SimCTG, a contrastive training objective to calibrate the model's representation space, and (ii) a decoding method-contrastive search-to encourage diversity while maintaining coherence in the generated text. Extensive experiments and analyses on three benchmarks from two languages demonstrate that our proposed approach significantly outperforms current state-of-the-art text generation methods as evaluated by both human and automatic metrics. © 2022, CC BY.

关键词： Decoding

Recovery of Distributed Iterative Solvers for Linear Systems Using Non-Volatile RAM

学校读者我要写书评

暂无评论

Recovery of Distributed Iterative Solvers for Linear Systems...

Fault Tolerance for HPC at eXtreme Scale (FTXS)

作者： Yehonatan Fridman Yaniv Snir Harel Levin Danny Hendler Hagit Attiya Gal Oren Department of Computer Science Ben-Gurion University of the Negev Israel Atomic Energy Commission Google Mobileye Vision Technologies Scientific Computing Center Nuclear Research Center – Negev Department of Computer Science Technion – Israel Institute of Technology

ISBN: (纸本)9781665488488

HPC systems are a critical resource for scientific research and advanced industries. The increased demand for computational power and memory ushers in the exascale era, in which supercomputers are designed to provide enormous computing power to meet these needs. These complex supercomputers consist of numerous compute nodes and are consequently expected to experience frequent faults and *** solvers, in particular, iterative linear solvers are key building block in numerous large-scale scientific applications. Consequently, supporting the recovery of distributed solvers is necessary for scaling scientific applications to exascale platforms. Previous recovery methods for iterative solvers are based on Checkpoint-Restart (CR), which incurs high fault tolerance overhead, or intrinsic fault tolerance, which require extra computation time to converge after *** state reconstruction (ESR) was proposed as an alternative mechanism to alleviate the impact of frequent failures on long-term computations. ESR has been shown to provide exact reconstruction of the computation state while avoiding the need for costly checkpointing. However, ESR currently relies on volatile memory for fault tolerance, and must therefore maintain redundancies in the RAM of multiple nodes. This not only incurs high memory overhead but also prevents ESR from being fully resilient, that is, resilient against a full system *** supercomputer designs feature emerging non-volatile RAM (NVRAM) technology, for example, the exascale Aurora that is planned to consist of Intel Optane™ DCPMM. This paper investigates how NVRAM can be utilized to devise an enhanced ESR-based recovery mechanism that is more efficient and provides full resilience. Our mechanism, called in-NVRAM ESR, provides full resiliency while significantly reducing both the memory footprint and the time overhead in comparison with the original ESR design (in-RAM ESR). In-NVRAM ESR is based on a novel MPI One-Si

关键词： Nonvolatile memory Source coding Fault tolerant systems Redundancy Random access memory Supercomputers computer crashes

Beyond Artificial Reality: Finding and Monitoring Live Events from Social Sensors

学校读者我要写书评

暂无评论

ACM Transactions on Internet technology 2020年第1期20卷 1-21页

作者： Pu, Calton Suprem, Abhijit Lima, Rodrigo Alves Musaev, Aibek Wang, De Irani, Danesh Webb, Steve Ferreira, Joao Eduardo School of Computer Science Georgia Institute of Technology Atlanta United States Department of Computer Science University of Alabama Tuscaloosa United States Sunmi US Inc United States Google United States Web Gnomes United States Department of Computer Science University of Sao Paulo Sao Paulo Brazil

With billions of active social media accounts and millions of live video cameras, live new big data offer many opportunities for smart applications. However, the main consumers of the new big data have been humans. We envision the research on live knowledge, to automatically acquire real-time, validated, and actionable information. Live knowledge presents two significant and diverging technical challenges: big noise and concept drift. We describe the EBKA (evidence-based knowledge acquisition) approach, illustrated by the LITMUS landslide information system. LITMUS achieves both high accuracy and wide coverage, demonstrating the feasibility and promise of EBKA approach to achieve live knowledge. © 2020 Association for Computing Machinery. All rights reserved.

关键词： Knowledge acquisition

The curious price of distributional robustness in reinforcement learning with a generative model 23

学校读者我要写书评

暂无评论

The curious price of distributional robustness in reinforcem...

Proceedings of the 37th International Conference on Neural Information Processing Systems

作者： Laixi Shi Gen Li Yuting Wei Yuxin Chen Matthieu Geist Yuejie Chi Department of Computing Mathematical Sciences California Institute of Technology CA and Caltech Department of Statistics The Chinese University of Hong Kong Hong Kong China and CUHK Department of Statistics and Data Science Wharton School University of Pennsylvania PA and UPenn Google Research Brain Team Paris France and Google Department of Electrical and Computer Engineering Carnegie Mellon University PA and CMU

This paper investigates model robustness in reinforcement learning (RL) via the framework of distributionally robust Markov decision processes (RMDPs). Despite recent efforts, the sample complexity of RMDPs is much less understood regardless of the uncertainty set in use; in particular, there exist large gaps between existing upper and lower bounds, and it is unclear if distributional robustness bears any statistical implications when benchmarked against standard RL. In this paper, assuming access to a generative model, we derive the sample complexity of RMDPs — when the uncertainty set is measured via either total variation or χ2 divergence over the full range of uncertainty levels — using a model-based algorithm called distributionally robust value iteration, and develop minimax lower bounds to benchmark its tightness. Our results not only strengthen the prior art in both directions of upper and lower bounds, but also deliver surprising messages that learning RMDPs is not necessarily easier or more difficult than standard MDPs. In the case of total variation, we establish the minimax-optimal sample complexity of RMDPs which is always smaller than that of standard MDPs. In the case of χ2 divergence, we establish the sample complexity of RMDPs that is tight up to polynomial factors of the effective horizon, and grows linearly with respect to the uncertainty level when it approaches infinity.

关键词：

Recovery of Distributed Iterative Solvers for Linear Systems Using Non-Volatile RAM

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Fridman, Yehonatan Snir, Yaniv Levin, Harel Hendler, Danny Attiya, Hagit Oren, Gal Department of Computer Science Ben-Gurion University of the Negev Israel Atomic Energy Commission Google Mobileye Vision Technologies Scientific Computing Center Nuclear Research Center - Negev Department of Computer Science Technion - Israel Institute of Technology

HPC systems are a critical resource for scientific research and advanced industries. The increased demand for computational power and memory ushers in the exascale era, in which supercomputers are designed to provide enormous computing power to meet these needs. These complex supercomputers consist of numerous compute nodes and are consequently expected to experience frequent faults and crashes. Mathematical solvers, in particular, iterative linear solvers are key building block in numerous large-scale scientific applications. Consequently, supporting the recovery of distributed solvers is necessary for scaling scientific applications to exascale platforms. Previous recovery methods for iterative solvers are based on Checkpoint-Restart (CR), which incurs high fault tolerance overhead, or intrinsic fault tolerance, which require extra computation time to converge after failures. Exact state reconstruction (ESR) was proposed as an alternative mechanism to alleviate the impact of frequent failures on long-term computations. ESR has been shown to provide exact reconstruction of the computation state while avoiding the need for costly checkpointing. However, ESR currently relies on volatile memory for fault tolerance, and must therefore maintain redundancies in the RAM of multiple nodes. This not only incurs high memory overhead but also prevents ESR from being fully resilient, that is, resilient against a full system crash. Recent supercomputer designs feature emerging non-volatile RAM (NVRAM) technology, for example, the exascale Aurora that is planned to consist of Intel Optane™ DCPMM. This paper investigates how NVRAM can be utilized to devise an enhanced ESR-based recovery mechanism that is more efficient and provides full resilience. Our mechanism, called in-NVRAM ESR, provides full resiliency while significantly reducing both the memory footprint and the time overhead in comparison with the original ESR design (in-RAM ESR). In-NVRAM ESR is based on a novel MPI One

关键词： Random access storage

Weighted QMIX: expanding monotonic value function factorisation for deep multi-agent reinforcement learning 20

学校读者我要写书评

暂无评论

Weighted QMIX: expanding monotonic value function factorisat...

Proceedings of the 34th International Conference on Neural Information Processing Systems

作者： Tabish Rashid Gregory Farquhar Bei Peng Shimon Whiteson Department of Computer Science University of Oxford Department of Computer Science University of Oxford and Google DeepMind

ISBN: (纸本)9781713829546

QMIX is a popular Q-learning algorithm for cooperative MARL in the centralised training and decentralised execution paradigm. In order to enable easy decentralisation, QMIX restricts the joint action Q-values it can represent to be a monotonic mixing of each agent's utilities. However, this restriction prevents it from representing value functions in which an agent's ordering over its actions can depend on other agents' actions. To analyse this representational limitation, we first formalise the objective QMIX optimises, which allows us to view QMIX as an operator that first computes the Q-learning targets and then projects them into the space repre-sentable by QMIX. This projection returns a representable Q-value that minimises the unweighted squared error across all joint actions. We show in particular that this projection can fail to recover the optimal policy even with access to Q*, which primarily stems from the equal weighting placed on each joint action. We rectify this by introducing a weighting into the projection, in order to place more importance on the better joint actions. We propose two weighting schemes and prove that they recover the correct maximal action for any joint action Q-values, and therefore for Q* as well. Based on our analysis and results in the tabular setting, we introduce two scalable versions of our algorithm, Centrally-Weighted (CW) QMIX and Optimistically-Weighted (OW) QMIX and demonstrate improved performance on both predator-prey and challenging multi-agent StarCraft benchmark tasks [26].

关键词：