Results 101 to 110 of about 31,169 (266)

Nonlinear permuted Granger causality

Canadian Journal of Statistics, EarlyView.
Abstract Granger causality is an established, contentious method that seeks causal temporal connections via association and precedence. While not true causal inference, it assists in mapping networks of information flow that may warrant further study.
Noah D. Gade, Jordan Rodu
wiley   +1 more source

Orchestrating Green Transformation: How AI Adoption Enables Corporate Carbon Neutrality

Corporate Social Responsibility and Environmental Management, EarlyView.
ABSTRACT As carbon neutrality has become a central goal of global climate governance, how firms achieve low‐carbon transformation has emerged as a critical research issue. However, prior studies have primarily focused on macro‐ or industry‐level analyses, offering limited and fragmented insights into how digital technologies—particularly AI—affect firm‐
Xiaonan Dong, Sungjin Son
wiley   +1 more source

Transformers are Multi-State RNNs

Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing
Transformers are considered conceptually different from the previous generation of state-of-the-art NLP models - recurrent neural networks (RNNs). In this work, we demonstrate that decoder-only transformers can in fact be conceptualized as unbounded multi-state RNNs - an RNN variant with unlimited hidden state size.
Matanel Oren, Michael Hassid, Yarden Nir, Yossi Adi, Roy Schwartz 0001   +4 more
openaire   +2 more sources

TAMNet: Temporal and adaptive‐frequency network with MixStyle for cross‐region oil and fluid production forecasting

Deep Underground Science and Engineering, EarlyView.
This paper presents temporal and adaptive‐frequency network with MixStyle (TAMNet), a deep time‐series modeling framework for accurate and robust multi‐well oil productivity forecasting. TAMNet integrates transformer and long short‐term memory architectures to capture both short‐ and long‐term temporal dependencies, enhanced by a temporal gate unit ...
Chunxi Yang, Leifeng Meng, Qinghu Fan, Ziang Guo, Ting Bao, Pingdong Zhong, Fulin Luo   +6 more
wiley   +1 more source

Artificial intelligence for adaptive neuromodulation in drug‐resistant epilepsy

Epilepsia, EarlyView.
Abstract Drug‐resistant epilepsy (DRE) affects nearly one third of people with epilepsy and is associated with substantial cognitive, psychiatric, and mortality burdens. For patients who are not candidates for resection or laser interstitial thermal therapy, neuromodulation therapies such as vagus nerve stimulation, deep brain stimulation, and ...
Amir Hossein Daraie, Arianna Damiani, Mahsa Khoshkhou, Luis A. Sanchez, Rachel J. Smith, Shruti H. Agashe, Jorge A. Gonzalez‐Martinez, Jennifer Hopp, Adam S. Charles, Sridevi V. Sarma, Joon‐Yi Kang   +10 more
wiley   +1 more source

AI‐based localization of the epileptogenic zone using intracranial EEG

Epilepsia Open, EarlyView.
Abstract Artificial intelligence (AI) is rapidly transforming our lives. Machine learning (ML) enables computers to learn from data and make decisions without explicit instructions. Deep learning (DL), a subset of ML, uses multiple layers of neural networks to recognize complex patterns in large datasets through end‐to‐end learning.
Atsuro Daida, Yipeng Zhang, Sotaro Kanai, Richard Staba, Vwani Roychowdhury, Hiroki Nariai   +5 more
wiley   +1 more source

On the Role of Depth in the Expressivity of RNNs

CoRR
The benefits of depth in feedforward neural networks are well known: composing multiple layers of linear transformations with nonlinear activations enables complex computations. While similar effects are expected in recurrent neural networks (RNNs), it remains unclear how depth interacts with recurrence to shape expressive power. Here, we formally show
Maude Lizaire, Michael Rizvi-Martel, Éric Dupuis, Guillaume Rabusseau   +3 more
openaire   +2 more sources

Regularizing RNNs by Stabilizing Activations

, 2015
We stabilize the activations of Recurrent Neural Networks (RNNs) by penalizing the squared distance between successive hidden states' norms. This penalty term is an effective regularizer for RNNs including LSTMs and IRNNs, improving performance on character-level language modeling and phoneme recognition, and outperforming weight noise and dropout.
David Krueger 0001, Roland Memisevic
openaire   +2 more sources

Artificial intelligence in preclinical epilepsy research: Current state, potential, and challenges

Epilepsia Open, EarlyView.
Abstract Preclinical translational epilepsy research uses animal models to better understand the mechanisms underlying epilepsy and its comorbidities, as well as to analyze and develop potential treatments that may mitigate this neurological disorder and its associated conditions. Artificial intelligence (AI) has emerged as a transformative tool across
Jesús Servando Medel‐Matus, Cesar Santana‐Gomez, Ruby G. Escalante, Dominique Duncan, Pedro F. Viana, Giulia Sofia Cereda, Naoto Kuroda, Aristea S. Galanopoulou   +7 more
wiley   +1 more source

Learning Rocking Dynamics From Sparse Shake‐Table Data With Interpretable Physics‐Informed Neural Networks

Earthquake Engineering &Structural Dynamics, EarlyView.
ABSTRACT We present a hybrid interpretable Physics‐Informed Neural Network Long‐Short Term Memory (Hybrid PINN LSTM) framework for predicting the seismic response of rocking blocks. Existing analytical models rely on uncertain idealizations, while purely data‐driven and machine‐learning approaches lack physical consistency and interpretability.
Shirley Shen, Christian Málaga‐Chuquitaype   +1 more
wiley   +1 more source