Results 81 to 90 of about 9,063 (254)

Numerical benchmarking of dual decomposition-based optimization algorithms for distributed model predictive control

Results in Control and Optimization
This paper presents a benchmark study of dual decomposition-based distributed optimization algorithms applied to constraint-coupled model predictive control problems.
Vassilios Yfantis, Achim Wagner, Martin Ruskowski   +2 more
doaj   +1 more source

Asynchronous Parallel Stochastic Quasi-Newton Methods. [PDF]

Parallel Comput, 2021
Tong Q, Liang G, Cai X, Zhu C, Bi J.
europepmc   +1 more source

Artificial intelligence empowers targeted protein degradation: Core technological innovations, multi‐scenario applications, and translational prospects

Smart Molecules, EarlyView.
AI is transforming TPD by improving the design, prediction, and optimization of degraders such as PROTACs, molecular glues, and LYTACs. This review summarizes key AI‐driven advances, highlights applications across drug discovery stages, and discusses remaining challenges and future directions for accelerating the development of therapies against ...
Shuanglin Qin, Rui Peng, Guangshuai Zhang, Si Yan, Jin Xiao, Zishu Liu, Ziyu Tan, Qingqing Xie, Shangjie Li, Wei Luo, Xiaohe Xiao   +10 more
wiley   +1 more source

New Results on Superlinear Convergence of Classical Quasi-Newton Methods. [PDF]

J Optim Theory Appl, 2021
Rodomanov A, Nesterov Y.
europepmc   +1 more source

A Novel Approach to Energy Management in Electric Steelworks

steel research international, EarlyView.
Feed‐forward neural networks are exploited to estimate electric energy consumptions of the electric arc furnace and ladle furnace processes. The models are used to optimize production schedule so that more energy intensive grades are produced when the cost of energy is lower.
Valentina Colla, Stefano Dettori, Silvia Cateni, Marco Vannucci, Antonella Vignali, Claudio Mocci, Irene Dovichi, Davide Chionna, Lorenzo Vannini, Enrico Paluzzano, Costanzo Pietrosanti, Davide Onesti, Daniele Venier   +12 more
wiley   +1 more source

A novel adaptive cubic quasi-Newton optimizer for deep learning based medical image analysis tasks, validated on detection of COVID-19 and segmentation for COVID-19 lung infection, liver tumor, and optic disc/cup. [PDF]

Med Phys, 2023
Liu Y, Zhang M, Zhong Z, Zeng X.
europepmc   +1 more source

Agglomeration Behavior of Nonmetallic Particles on the Surface of Si‐Killed Liquid Steel Using HT‐CSLM: An In Situ Investigation

steel research international, EarlyView.
The in situ agglomeration behavior of semisolid Al–silicate inclusions in Si‐killed steel is studied using high‐temperature confocal scanning laser microscopy. Particle motion and interparticle forces at the steel/argon interface are quantified and compared with a modified capillary interaction model.
Suwam Kumar, Muhammad Nabeel, Angshuman Podder, Andre B. Phillion, Neslihan Dogan   +4 more
wiley   +1 more source

Computing the Projection on the Intersection of Linear Equality and Cardinality Constraints

IEEE Access
In this paper, we introduce a novel approach to solve the vector projection problem under linear equality and cardinality constraints, termed the Quasi-Newton Lagrangian method. This approach employs a Lagrangian transformation to address linear equality
Xiao-Jing Wang, Zhen-Sheng Yu
doaj   +1 more source

Sensitivity study on accuracy and relevant parameters in numerical modeling of internal tendon breaks

Structural Concrete, EarlyView.
Abstract This study investigates the numerical modeling of internal tendon breaks in prestressed concrete structures and evaluates the key parameters affecting accuracy. A combined approach is followed, involving global sensitivity analysis via the Elementary Effects Method and subsequent model calibration using experimental data from distributed fiber‐
Aeneas Paul, David Sanio, Peter Mark
wiley   +1 more source

Pre‐task Vocabulary Support Enhances Lexical Learning but Dampens Positive Emotions: Interactive Task Implementation in English‐as‐a‐Foreign‐Language Classroom

TESOL Quarterly, EarlyView.
Abstract This classroom study explored the effects of vocabulary support on collocation learning and affective responses in task‐based language teaching (TBLT) among English‐as‐a‐foreign‐language (EFL) learners at a Japanese university. For this purpose, 68 EFL learners completed two interactive information‐gap tasks under either vocabulary‐support or ...
Yuichi Suzuki, Sachiko Nakamura
wiley   +1 more source