José I. Mestre

16 papers C 2Journal 12Unranked 2

Year	Rank	Type	Title / Venue / Authors
2026	J	`jnl`	Regime Change Hypothesis: Foundations for Decoupled Dynamics in Neural Network Training. CoRR Cristian Pérez-Corral, Alberto Fernández-Hernández, José I. Mestre, Manuel F. Dolz, José Duato, Enrique S. Quintana-Ortí
2026	J	`jnl`	When Learning Rates Go Wrong: Early Structural Signals in PPO Actor-Critic. CoRR Alberto Fernández-Hernández, Cristian Pérez-Corral, José I. Mestre, Manuel F. Dolz, José Duato, Enrique S. Quintana-Ortí
2026	J	`jnl`	Why Adam Works Better with β CoRR Alberto Fernández-Hernández, Cristian Pérez-Corral, José I. Mestre, Manuel F. Dolz, Enrique S. Quintana-Ortí
2026	J	`jnl`	λ-GELU: Learning Gating Hardness for Controlled ReLU-ization in Deep Networks. CoRR Cristian Pérez-Corral, Alberto Fernández-Hernández, José I. Mestre, Manuel F. Dolz, Enrique S. Quintana-Ortí
2025	—	`conf`	Decoupling Structural and Quantitative Knowledge in ReLU-based Deep Neural Networks. EuroMLSys José Duato, José I. Mestre, Manuel F. Dolz, Enrique S. Quintana-Ortí, José Cano
2025	J	`jnl`	Deep learning inference optimisation for IoT: Conv2D-ReLU-BN layer fusion and quantisation. J. Supercomput. José I. Mestre, Sergio Barrachina, Darwin Quezada, Manuel F. Dolz
2025	J	`jnl`	GLAI: GreenLightningAI for Accelerated Training through Knowledge Decoupling. CoRR José I. Mestre, Alberto Fernández-Hernández, Cristian Pérez-Corral, Manuel F. Dolz, José Duato, Enrique S. Quintana-Ortí
2025	J	`jnl`	Latency-Critical Quantized Inference With Transformer Decoders on ARM and RISC-V CPUs. IEEE Internet Things J. Héctor Martínez, Sandra Catalán, Adrián Castelló, José I. Mestre, Enrique S. Quintana-Ortí
2025	J	`jnl`	OUI Need to Talk About Weight Decay: A New Perspective on Overfitting Detection. CoRR Alberto Fernández-Hernández, José I. Mestre, Manuel F. Dolz, José Duato, Enrique S. Quintana-Ortí
2025	J	`jnl`	Sinusoidal Initialization, Time for a New Start. CoRR Alberto Fernández-Hernández, José I. Mestre, Manuel F. Dolz, José Duato, Enrique S. Quintana-Ortí
2023	J	`jnl`	GreenLightningAI: An Efficient AI System with Decoupled Structural and Quantitative Knowledge. CoRR José Duato, José I. Mestre, Manuel F. Dolz, Enrique S. Quintana-Ortí
2023	J	`jnl`	Using machine learning to model the training scalability of convolutional neural networks on clusters of GPUs. Computing Sergio Barrachina, Adrián Castelló, Mar Catalán, Manuel F. Dolz, José I. Mestre
2021	—	`conf`	A Flexible Research-Oriented Framework for Distributed Training of Deep Neural Networks. IPDPS Workshops Sergio Barrachina, Adrián Castelló, Mar Catalán, Manuel F. Dolz, José I. Mestre
2021	C	`conf`	Evaluation of MPI Allreduce for Distributed Training of Convolutional Neural Networks. PDP Adrián Castelló, Mar Catalán, Manuel F. Dolz, José I. Mestre, Enrique S. Quintana-Ortí, José Duato
2021	C	`conf`	Performance Modeling for Distributed Training of Convolutional Neural Networks. PDP Adrián Castelló, Mar Catalán, Manuel F. Dolz, José I. Mestre, Enrique S. Quintana-Ortí, José Duato
2021	J	`jnl`	PyDTNN: A user-friendly and extensible framework for distributed deep learning. J. Supercomput. Sergio Barrachina, Adrián Castelló, Mar Catalán, Manuel F. Dolz, José I. Mestre