Cristian Pérez-Corral

According to our database, Cristian Pérez-Corral authored at least 13 papers between 2025 and 2026.

Collaborative distances:

• Dijkstra number of four.
• Erdős number of four.

2026

Detecting Atypical Clients in Federated Learning via Representation-Level Divergence.

Cristian Pérez-Corral, José I. Mestre, Alberto Fernández-Hernández, Manuel F. Dolz, Enrique S. Quitana-Ortí

CoRR, May, 2026

StableGrad: Backward Scale Control without Batch Normalization.

José I. Mestre, Alberto Fernández-Hernández, Cristian Pérez-Corral, Manuel F. Dolz, Enrique S. Quintana-Ortí

CoRR, May, 2026

FedOUI: OUI-Guided Client Weighting for Federated Aggregation.

Alberto Fernández-Hernández, José I. Mestre, Cristian Pérez-Corral, Manuel F. Dolz, José Duato, Enrique S. Quintana-Ortí

CoRR, May, 2026

OUI as a Structural Observable: Towards an Activation-Centric View of Neural Network Training.

Alberto Fernández-Hernández, José I. Mestre, Cristian Pérez-Corral, Manuel F. Dolz, José Duato, Enrique S. Quintana-Ortí

CoRR, May, 2026

OUIDecay: Adaptive Layer-wise Weight Decay for CNNs Using Online Activation Patterns.

Alberto Fernández-Hernández, José I. Mestre, Cristian Pérez-Corral, Manuel F. Dolz, José Duato, Enrique S. Quintana-Ortí

CoRR, May, 2026

Refresh-Scaling the Memory of Balanced Adam.

Alberto Fernández-Hernández, Cristian Pérez-Corral, José I. Mestre, Manuel F. Dolz, Enrique S. Quintana-Ortí

CoRR, May, 2026

FedSQ: Optimized Weight Averaging via Fixed Gating.

Cristian Pérez-Corral, José I. Mestre, Alberto Fernández-Hernández, Manuel F. Dolz, José Duato, Enrique S. Quintana-Ortí

CoRR, April, 2026

λ-GELU: Learning Gating Hardness for Controlled ReLU-ization in Deep Networks.

Cristian Pérez-Corral, Alberto Fernández-Hernández, José I. Mestre, Manuel F. Dolz, Enrique S. Quintana-Ortí

CoRR, March, 2026

When Learning Rates Go Wrong: Early Structural Signals in PPO Actor-Critic.

Alberto Fernández-Hernández, Cristian Pérez-Corral, José I. Mestre, Manuel F. Dolz, José Duato, Enrique S. Quintana-Ortí

CoRR, March, 2026

Regime Change Hypothesis: Foundations for Decoupled Dynamics in Neural Network Training.

Cristian Pérez-Corral, Alberto Fernández-Hernández, José I. Mestre, Manuel F. Dolz, José Duato, Enrique S. Quintana-Ortí

CoRR, February, 2026

Why Adam Works Better with β₁ = β₂: The Missing Gradient Scale Invariance Principle.

Alberto Fernández-Hernández, Cristian Pérez-Corral, José I. Mestre, Manuel F. Dolz, Enrique S. Quintana-Ortí

CoRR, January, 2026

2025

GLAI: GreenLightningAI for Accelerated Training through Knowledge Decoupling.

José I. Mestre, Alberto Fernández-Hernández, Cristian Pérez-Corral, Manuel F. Dolz, José Duato, Enrique S. Quintana-Ortí

CoRR, October, 2025

From domain-landmark graph learning to problem-landmark graph generation.

Cristian Pérez-Corral, Antonio Garrido, Laura Sebastia

CoRR, September, 2025