检索结果-内蒙古大学图书馆

Coordinated management of processor configuration and cache partitioning to optimize energy under QoS constraints

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Nejat, Mehrzad Manivannan, Madhavan Pericàs, Miquel Stenström, Per Department of Computer Science and Engineering Chalmers University of Technology

An effective way to improve energy efficiency is to throttle hardware resources to meet a certain performance target, specified as a QoS constraint, associated with all applications running on a multicore system. Prior art has proposed resource management (RM) frameworks in which the share of the last-level cache (LLC) assigned to each processor and the voltage-frequency (VF) setting for each processor is managed in a coordinated fashion to reduce energy. A drawback of such a scheme is that, while one core gives up LLC resources for another core, the performance drop must be compensated by a higher VF setting which leads to a quadratic increase in energy consumption. By allowing each core to be adapted to exploit instruction and memory-level parallelism (ILP/MLP), substantially higher energy savings are enabled. This paper proposes a coordinated RM for LLC partitioning, processor adaptation, and per-core VF scaling. A first contribution is a systematic study of the resource trade-offs enabled when trading between the three classes of resources in a coordinated fashion. A second contribution is a new RM framework that utilizes these trade-offs to save more energy. Finally, a challenge to accurately model the impact of resource throttling on performance is to predict the amount of MLP with high accuracy. To this end, the paper contributes with a mechanism that estimates the effect of MLP over different processor configurations and LLC allocations. Overall, we show that up to 18% of energy, and on average 10%, can be saved using the proposed scheme. Copyright © 2019, The Authors. All rights reserved.

关键词： Resource allocation

Corrigendum to “Modeling PROTAC degradation activity with machine learning” [Artificial Intelligence in the Life sciences 6 (2024) 100104]

学校读者我要写书评

暂无评论

Artificial Intelligence in the Life sciences 2024年 6卷

作者： Stefano Ribes Eva Nittinger Christian Tyrchan Rocío Mercado Department of Computer Science and Engineering Section for Data Science and AI Chalmers University of Technology Chalmersplatsen 4 Gothenburg 412 96 Sweden Medicinal Chemistry Research and Early Development Respiratory and Immunology (R&I) BioPharmaceuticals R&D AstraZeneca Pepparedsleden 1 Mölndal 431 83 Sweden

PROTACs are a promising therapeutic modality that harnesses the cell’s built-in degradation machinery to degrade specific proteins. Despite their potential, developing new PROTACs is challenging and requires significant domain expertise, time, and cost. Meanwhile, machine learning has transformed drug design and development. In this work, we present a strategy for curating open-source PROTAC data and an open-source deep learning tool for predicting the degradation activity of novel PROTAC molecules. The curated dataset incorporates important information such as p D C 50 , D m a x , E3 ligase type, POI amino acid sequence, and experimental cell type. Our model architecture leverages learned embeddings from pretrained machine learning models, in particular for encoding protein sequences and cell type information. We assessed the quality of the curated data and the generalization ability of our model architecture against new PROTACs and targets via three tailored studies, which we recommend other researchers to use in evaluating their degradation activity models. In each study, three models predict protein degradation in a majority vote setting, reaching a top test accuracy of 80.8% and 0.865 ROC-AUC, and a test accuracy of 62.3% and 0.604 ROC-AUC when generalizing to novel protein targets. Our results are not only comparable to state-of-the-art models for protein degradation prediction, but also part of an open-source implementation which is easily reproducible and less computationally complex than existing approaches.

关键词： PROTAC Machine learning Drug discovery Protein degradation

Justicia: A stochastic SAT approach to formally verify fairness

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Ghosh, Bishwamittra Basu, Debabrota Meel, Kuldeep S. School of Computing National University of Singapore Singapore Department of Computer Science and Engineering Chalmers University of Technology Göteborg Sweden Scool Inria Lille-Nord Europe France

As a technology ML is oblivious to societal good or bad, and thus, the field of fair machine learning has stepped up to propose multiple mathematical definitions, algorithms, and systems to ensure different notions of fairness in ML applications. Given the multitude of propositions, it has become imperative to formally verify the fairness metrics satisfied by different algorithms on different datasets. In this paper, we propose a stochastic satisfiability (SSAT) framework, Justicia, that formally verifies different fairness measures of supervised learning algorithms with respect to the underlying data distribution. We instantiate Justicia on multiple classification and bias mitigation algorithms, and datasets to verify different fairness metrics, such as disparate impact, statistical parity, and equalized odds. Justicia is scalable, accurate, and operates on non-Boolean and compound sensitive attributes unlike existing distribution-based verifiers, such as FairSquare and VeriFair. Being distribution-based by design, Justicia is more robust than the verifiers, such as AIF360, that operate on specific test samples. We also theoretically bound the finite-sample error of the verified fairness measure. © 2020, CC BY-NC-SA.

关键词： Stochastic systems

Software Logs for Machine Learning in a DevOps Environment

学校读者我要写书评

暂无评论

Software Logs for Machine Learning in a DevOps Environment

Euromicro Conference on Software engineering and Advanced Applications (SEAA)

作者： Nathan Bosch Jan Bosch Development Unit Radio Services & Infra Analytics (DSI Analytics) Ericsson AB Gothenburg Sweden Department of Computer Science and Engineering Chalmers University of Technology Gothenburg Sweden

ISBN: (数字)9781728195322

ISBN: (纸本)9781728195339

System logs perform a critical function in software-intensive systems as logs record the state of the system and significant events in the system at important points in time. Unfortunately, log entries are typically created in an ad-hoc, unstructured and uncoordinated fashion, limiting their usefulness for analytics and machine learning. In a DevOps environment, especially, unmanaged evolution in log data structure causes frequent disruption of operations in automated data pipelines, dashboards and analytics. In this paper, we present the main challenges of contemporary approaches to generating, storing and managing the evolution of system logs data for large, complex, software-intensive systems based on an in-depth case study at a world-leading telecommunications company. Second, we present an approach for generating and managing the evolution of log data in a DevOps environment that does not suffer from the aforementioned challenges and is optimized for use in machine learning. Third, we provide validation of the approach based on expert interviews that confirm that the approach addresses the identified challenges and problems.

关键词： Machine learning Companies Software Research and development Standards Data models Training

Using Valued Booleans to Find Simpler Counterexamples in Random Testing of Cyber-Physical Systems

学校读者我要写书评

暂无评论

IFAC-PapersOnLine 2018年第7期51卷 408-415页

作者： Claessen, Koen Smallbone, Nicholas Eddeland, Johan Ramezani, Zahra Åkesson, Knut Department of Computer Science and Engineering Chalmers University of Technology Gothenburg Sweden Department of Electrical Engineering Chalmers University of Technology Gothenburg Sweden Volvo Car Corporation Gothenburg Sweden

We propose a new logic of valued Booleans for writing properties which are not just true or false but compute how severely they are falsified. The logic is reminiscent of STL or MTL but gives the tester control over what severity means in the particular problem domain. We use this logic to simplify failing test inputs in the context of random testing of cyber-physical systems and show that it improves the quality of counterexamples found. The logic of valued Booleans might also be used as an alternative to the standard robust semantics of STL formulas in optimization-based approaches to falsification. © 2018

关键词： Embedded systems Applications computer circuits computer control systems Control system analysis Cyber Physical System Hybrid systems Semantics Temporal logic Testing Verification Embedded computer control systems Embedded computer Systems Logical design Model driven Physical design Reachability analysis supervision

Error mitigation via stabilizer measurement emulation

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Greene, Ami Kjaergaard, M. Schwartz, M.E. Samach, G.O. Bengtsson, A. McNally, C. O’Keeffe, M. Kim, D.K. Marvian, M. Melville, A. Niedzielski, B.M. Vepsäläinen, A. Winik, R. Yoder, J. Rosenberg, D. Lloyd, S. Orlando, T.P. Marvian, I. Gustavsson, S. Oliver, W.D. Research Laboratory of Electronics Massachusetts Institute of Technology CambridgeMA02139 United States MIT Lincoln Laboratory LexingtonMA02421 United States Microtechnology and Nanoscience Chalmers University of Technology GöteborgSE-412 96 Sweden Department of Mechanical Engineering Massachusetts Institute of Technology CambridgeMA02139 United States Departments of Physics & Electrical and Computer Engineering Duke University DurhamNC27708 United States Department of Physics Massachusetts Institute of Technology CambridgeMA02139 United States Department of Electrical Engineering & Computer Science Massachusetts Institute of Technology CambridgeMA02139 United States

Dynamical decoupling (DD) is a widely-used quantum control technique that takes advantage of temporal symmetries in order to partially suppress quantum errors without the need resource-intensive error detection and correction protocols. This and other open-loop error mitigation techniques are critical for quantum information processing in the era of Noisy Intermediate-Scale Quantum technology. However, despite its utility, dynamical decoupling does not address errors which occur at unstructured times during a circuit, including certain commonly-encountered noise mechanisms such as cross-talk and imperfectly calibrated control pulses. Here, we introduce and demonstrate an alternative technique — ‘quantum measurement emulation’ (QME) — that effectively emulates the measurement of stabilizer operators via stochastic gate application, leading to a first-order insensitivity to coherent errors. The QME protocol enables error suppression based on the stabilizer code formalism without the need for costly measurements and feedback, and it is particularly well-suited to discrete coherent errors that are challenging for DD to address. Copyright © 2021, The Authors. All rights reserved.

关键词： Errors

Human exposure to aerosol from indoor gas stove cooking and the resulting cardiovascular system responses

学校读者我要写书评

暂无评论

Toxicology Reports 2024年 13卷 101716页

作者： Naseri, Motahareh Sultanbekovna, Aigerim Abilova Malekipirbazari, Milad Kenzhegaliyeva, Elzira Buonanno, Giorgio Stabile, Luca Hopke, Philip K. Cassee, Flemming Crape, Byron Sabanov, Sergei Zhumambayeva, Saule Ozturk, Fatma Tadi, Mehrdad Jafari Torkmahalleh, Mehdi Amouei Shah, Dhawal Department of Chemical and Materials Engineering School of Engineering and Digital Sciences Nazarbayev University Astana Kazakhstan Clinical Academic Department of Laboratory Medicine Pathology and Genetics University Medical Center Astana Kazakhstan Department of Computer Science and Engineering Chalmers University of Technology Gothenburg SE41296 Sweden Department of Civil and Mechanical Engineering University of Cassino and Southern Lazio via Di Biasio 43 Cassino 03043 Italy Department of Public Health Sciences University of Rochester Medical Center Rochester 14642 NY United States National Institute for Public Health and the Environment Bilthoven Netherlands Department of Medicine School of Medicine Nazarbayev University Astana Kazakhstan Department of mining School of Mining and Geosciences Nazarbayev University Astana Kazakhstan Department of children diseases with courses in allergy hematology and endocrinology Astana Medical University Astana Kazakhstan Environmental Engineering Department Faculty of Engineering Bolu Abant Izzet Baysal University (BAIBU) Golkoy Campus Bolu 14030 Turkey Department of Cell and Molecular Medicine Rush University Medical Center Chicago 60607 IL United States Division of Environmental and Occupational Health Sciences School of Public Health University of Illinois at Chicago Chicago 60612 IL United States

The effect of cooking aerosol on the human heart was investigated in this study. The heart rate and blood pressure of 33 healthy adults were monitored before, exactly after, and two hours post-exposure (30 minutes, 60 minutes, 90 minutes, and 120 minutes after cooking). One hundred twenty grams of ground beef was fried in sunflower oil for twenty minutes using a gas stove without ventilation. Ultrafine particles, indoor temperature, relative humidity, carbon dioxide, oil, and meat temperatures were monitored during the experiment. The average particle emission rate (S) and average decay rate (a+k) for meat frying were found to be 2.09×10¹³ (SD=3.94 ×10¹³, R²=0.98, P <0.0001) particles/min, and 0.055 (SD=0.019, R²=0.91, P <0.0001) particles/min, respectively. No statistically significant changes in diastolic blood pressure (DBP) and heart rate (HR) were observed. The average systolic blood pressure (SBP) statistically significantly increased from 98 mmHg (before the exposure) to 106 mmHg 60 minutes after the exposure. The results suggested that frying emission statistically significantly impacted blood pressure. © 2024 The Authors

关键词： Diastolic blood pressure Frying aerosol Heart rate Systolic blood pressure Ultrafine particles

Mena compared to europe: The influence of land use, nuclear power, and transmission expansion on renewable electricity system costs

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Ek Fälth, H. Atsmon, D. Reichenberg, L. Verendel, V. Department of Space Earth and Environment Chalmers University of Technology Göteborg412 96 Sweden Department of Mathematics and Systems Analysis Aalto University Otakaari 1 F Espoo Finland Department of Computer Science and Engineering Chalmers University of Technology Göteborg412 96 Sweden

Most studies of near-zero-carbon power systems consider Europe and the United States. In this paper, we focus on the Middle East and North Africa (MENA), where weather conditions, especially for solar, differ substantially from those in Europe. We use a green-field linear capacity expansion model with over-night investment to assess the effect on system cost of (i) limiting/expanding the amount of land available for wind and solar farms, (ii) allowing for nuclear power and (iii) disallowing for international transmission. This is done under three different cost regimes for solar PV and battery storage. We find that: • The amount of available land for wind and solar farms can have a great impact on the system cost. We found a cost increase of 0-50% as a result of reduced available land. In MENA, the impact on system cost is greatly influenced by the PV and battery cost regime, which is not the case in Europe. • Allowing for nuclear has nearly no effect in MENA, while it can decrease system costs in Europe by up to 23%. In Europe, the effect on system cost of whether nuclear power is allowed is highly dependent on the PV and battery cost regime, which is not the case in MENA. • Disallowing for international transmission increases costs by up to around 25% in both Europe and MENA. The cost increase depends on cost regime for PV and batteries. The impact on system cost off these three controversial parts of a decarbonized power system thus plays out differently, depending on (i) the region and (ii) uncertain future costs for solar PV and storage. We conclude that a renewable power system in MENA, is less costly than in Europe irrespective of the cost regime. In MENA, the system costs vary between 37 and 83 e/MWh. In Europe, the system costs vary between 43 and 89 e/MWh. Copyright © 2020, The Authors. All rights reserved.

关键词： Land use

Traction adaptive motion planning and control at the limits of handling

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Svensson, Lars Bujarbaruah, Monimoy Karsolia, Arpit Berger, Christian Törngren, Martin Department of Machine Design KTH Royal Institute of Technology Sweden Model Predictive Control Laboratory University of California Berkeley United States Department of Electrical Engineering Chalmers University of Technology Sweden Department of Computer Science and Engineering University of Gothenburg Sweden

In this paper, we address the problem of motion planning and control at the limits of handling, under locally varying traction conditions. We propose a novel solution method where traction variations over the prediction horizon are represented by time-varying tire force constraints, derived from a predictive friction estimate. A constrained finite time optimal control problem is solved in a receding horizon fashion, imposing these time-varying constraints. Furthermore, our method features an integrated sampling augmentation procedure that addresses the problems of infeasibility and sensitivity to local minima that arise at abrupt constraint alterations, e.g., due to sudden friction changes. We validate the proposed algorithm on a Volvo FH16 heavy-duty vehicle, in a range of critical scenarios. Experimental results indicate that traction adaptive motion planning and control improves the vehicle's capacity to avoid accidents, both when adapting to low local traction, by ensuring dynamic feasibility of the planned motion, and when adapting to high local traction, by realizing high traction utilization. Copyright © 2020, The Authors. All rights reserved.

关键词： Motion planning