Results 71 to 80 of about 5,803,394 (295)

Decoding Structure‐Property Relationships in Anion Exchange Membranes via a Chemically Informed Dual‐Channel Graph Attention Network

Advanced Science, EarlyView.
SPARK decodes structure‐property relationships in anion exchange membranes (AEMs) via a chemically informed dual‐channel graph attention network (DEGAT) that explicitly captures microphase separation. It outputs five‐level grades for hydroxide conductivity and alkaline stability and highlights relevant key structural units, enabling robust pre ...
Wanting Chen, Ye Hu, Zijun Xiao, Deming Xia, Bo Pang, Xuemei Wu, Gaohong He   +6 more
wiley   +1 more source

Cost-Effective Multitask Active Learning in Wearable Sensor Systems

Sensors
Multitask learning models provide benefits by reducing model complexity and improving accuracy by concurrently learning multiple tasks with shared representations.
Asiful Arefeen, Hassan Ghasemzadeh
doaj   +1 more source

Lifelong Personalization via Gaussian Process Modeling for Long-Term HRI

Frontiers in Robotics and AI, 2021
Across a wide variety of domains, artificial agents that can adapt and personalize to users have potential to improve and transform how social services are provided.
Samuel Spaulding, Jocelyn Shen, Hae Won Park, Cynthia Breazeal   +3 more
doaj   +1 more source

GPTune: multitask learning for autotuning exascale applications

ACM SIGPLAN Symposium on Principles & Practice of Parallel Programming, 2021
Multitask learning has proven to be useful in the field of machine learning when additional knowledge is available to help a prediction task. We adapt this paradigm to develop autotuning frameworks, where the objective is to find the optimal performance ...
Yang Liu, Wissam M. Sid-Lakhdar, O. Marques, Xinran Zhu, Chang Meng, J. Demmel, X. Li   +6 more
semanticscholar   +1 more source

WS2 Optoelectronic Memristive Reservoir Enabling Ultra‐Low‐Power, Multi‐Task, and Environmentally Stable Neuromorphic Computing

Advanced Science, EarlyView.
WS2‐based in‐memory sensing reservoir computing integrates sensing, memory, and computation in one compact device. It achieves ∼94% N‐MNIST, ∼93% eye motion perception, and ∼89% speech recognition with ultra‐low energy (∼25.5 fJ/spike). The system shows stability at 95% humidity, endurance over 1.5M cycles, and supports synaptic plasticity, enabling ...
Dayanand Kumar, Hanrui Li, Divyanshu Divyanshu, Dhananjay D. Kumbhar, Manoj Kumar Rajbhar, Amit Singh, Abdul Momin Syed, Selma Amara, Gianluca Setti, Nazek El‐Atab   +9 more
wiley   +1 more source

Multitask Learning Based on Least Squares Support Vector Regression for Stock Forecast

Axioms, 2022
Various factors make stock market forecasting difficult and arduous. Single-task learning models fail to achieve good results because they ignore the correlation between multiple related tasks. Multitask learning methods can capture the cross-correlation
Heng-Chang Zhang, Qing Wu, Fei-Yan Li, Hong Li   +3 more
doaj   +1 more source

Consistent Multitask Learning with Nonlinear Output Relations [PDF]

, 2017
Key to multitask learning is exploiting relationships between different tasks to improve prediction performance. If the relations are linear, regularization approaches can be used successfully.
Ciliberto, Carlo, Pontil, Massimiliano, Rosasco, Lorenzo, Rudi, Alessandro   +3 more
core   +2 more sources

Adaptation and learning over networks for nonlinear system modeling

, 2017
In this chapter, we analyze nonlinear filtering problems in distributed environments, e.g., sensor networks or peer-to-peer protocols. In these scenarios, the agents in the environment receive measurements in a streaming fashion, and they are required to
Argyriou, Balakrishnan, Bouboulis, Bouboulis, Cattivelli, Cevher, Chen, Chen, Chen, Chen, Chen, Chen, Chen, Chen, Chouvardas, Di Lorenzo, Di Lorenzo, Evgeniou, Forero, Gao, Honeine, Honeine, Huang, Igelnik, Jin, Lazarevic, Li, Lopes, Mateos, Matta, Nassif, Nassif, Navia-Vazquez, Parreira, Predd, Rahimi, Richard, Rusu, Sandryhaila, Sayed, Sayed, Scardapane, Scardapane, Scardapane, Scardapane, Scarpiniti, Scarpiniti, Shin, Singh, Tsitsiklis, Yuan, Zhao, Zhao   +52 more
core   +2 more sources

Enhanced High Dimensionality and the Information Processing Capacity in Interfered Spin Wave‐Based Reservoir Computing, Achieved With Eight Detectors

Advanced Electronic Materials, EarlyView.
Physical reservoir computing (PRC) based on spin wave interference has demonstrated high computational performance, yet room for improvement remains. In this study, we fabricated this concept PRC with eight detectors and evaluated the impact of the number of detectors using a chaotic time series prediction task.
Sota Hikasa, Wataru Namiki, Daiki Nishioka, Maki Nishimura, Ryo Iguchi, Kazuya Terabe, Takashi Tsuchiya   +6 more
wiley   +1 more source

Geolocation with Attention-Based Multitask Learning Models [PDF]

Proceedings of the 5th Workshop on Noisy User-generated Text (W-NUT 2019), 2019
Geolocation, predicting the location of a post based on text and other information, has a huge potential for several social media applications. Typically, the problem is modeled as either multi-class classification or regression. In the first case, the classes are geographic areas previously identified; in the second, the models directly predict ...
Fornaciari, Tommaso, Hovy, Dirk
openaire   +2 more sources