Results 91 to 100 of about 14,848 (289)

Cascaded channel pruning using hierarchical self-distillation

Proceedings of the British Machine Vision Conference 2020, 2020
In this paper, we propose an approach for filter-level pruning with hierarchical knowledge distillation based on the teacher, teaching-assistant, and student framework. Our method makes use of teaching assistants at intermediate pruning levels that share the same architecture and weights as the target student.
Miles, Roy, Mikolajczyk, Krystian
openaire   +2 more sources

Coacervates Made of Elastin‐Like Polypeptides Fused with Melanocyte‐Stimulating Hormone and Monocyte Chemoattractant Protein Enhance Skin Wound Healing in Spinal Cord‐Injured Mice

Advanced Healthcare Materials, EarlyView.
Pressure skin wounds are frequent complications after spinal cord injury (SCI), with impaired healing due to vascular and immune deficits. Elastin‐like polypeptides (ELP) fused to α‐MSH (MSH‐ELP) or MCP‐1 (MCP‐ELP) are developed and tested on these wounds. The resulting nanoparticles are non‐toxic and bioactive, and they enhance macrophage recruitment,
Suneel Kumar, Dhruv Patel, Greeshma Manjunath, Shashank Madhavan, Ashley Mathew, Dnyaneshwari Rananavare, Jahnavi Shah, Francois Berthiaume   +7 more
wiley   +1 more source

Learning from All Sides: Diversified Positive Augmentation via Self-distillation in Recommendation

, 2023
Personalized recommendation relies on user historical behaviors to provide user-interested items, and thus seriously struggles with the data sparsity issue.
Lin, Leyu, Liu, Chong, Liu, Xiaoyang, Xia, Feng, Xie, Ruobing, Zhang, Lixin   +5 more
core  

Self-Distillation Amplifies Regularization in Hilbert Space

, 2020
Knowledge distillation introduced in the deep learning context is a method to transfer knowledge from one architecture to another. In particular, when the architectures are identical, this is called self-distillation. The idea is to feed in predictions of the trained model as new target values for retraining (and iterate this loop possibly a few times).
Mobahi, Hossein, Farajtabar, Mehrdad, Bartlett, Peter L.   +2 more
openaire   +2 more sources

Dual‐Peptide Nanoplatform: Mesoporous Silica Nanoparticles Functionalized With a Cell‐Penetrating Peptide and Loaded With Rationally Designed Antimicrobial Peptides for Tuberculosis Therapy

Advanced Healthcare Materials, EarlyView.
Machine learning–guided engineering of a plectasin‐derived peptide yields DC05, a potent antimycobacterial candidate. Encapsulation into tuftsin‐functionalized mesoporous silica nanoparticles enhances intracellular delivery, stability, and activity against Mycobacterium tuberculosis while maintaining low cytotoxicity and minimal hemolysis. The combined
Christian S. Carnero Canales, Cesar Augusto Roque‐Borda, Ana Carolina Cerqueira Negri, Oswaldo Ramirez Delgado, Letícia Oliveira Catarin Nunes, Flávia Aparecida Resende, Hernane Barud, Norival Alves Santos‐Filho, Saulo Santesso Garrido, Karen Cristina Oliveira, Alexandra Ivo de Medeiros, Rafael Miguel Sábio, Fernando Rogério Pavan   +12 more
wiley   +1 more source

Organelle Localization‐Induced Bio‐Orthogonal Polymerization (OLIBOP) for Photostable Super‐Resolution Live‐Cell Imaging

Advanced Healthcare Materials, EarlyView.
Organelle localization‐induced biorthogonal polymerization enables direct synthesis of photostable poly‐AIEgens within targeted organelles for super‐resolution live‐cell imaging. ABSTRACT Real‐time monitoring of dynamic biological processes demands fluorescent probes that can withstand prolonged light exposure without photobleaching—a critical ...
Gaeun Park, Sangpil Kim, Dohyun Kim, Injun Hwang, Ja‐Hyoung Ryu   +4 more
wiley   +1 more source

Semantic Super-Resolution via Self-Distillation and Adversarial Learning

IEEE Access
Semantic super-resolution (SR) is an approach that improves the SR performance by leveraging semantic information about the scene. This study develops a novel semantic SR method that is based on the generative adversarial network (GAN) framework and self-
Hanhoon Park
doaj   +1 more source

Federated Learning on Heterogeneous Data via Adaptive Self-Distillation

, 2023
Federated Learning (FL) is a machine learning paradigm that enables clients to jointly train a global model by aggregating the locally trained models without sharing any local training data. In practice, there can often be substantial heterogeneity (e.g.,
Chakraborty, Anirban, Nayak, Gaurav Kumar, Simmhan, Yogesh, Singh, Arya, Yashwanth, M.   +4 more
core  

Optoelectrical Devices for Neural Interfacing: Engineering Integration, Stability, and Multimodal Sensing

Advanced Healthcare Materials, EarlyView.
Implantable optoelectrical devices are an effective resource for the modulation and monitoring of neural activity with high spatiotemporal resolution. This review discusses current challenges faced by these devices and outlines future perspectives for the development of next‐generation neural interfaces targeting chronic, multisite, and multimodal ...
Stella Aslanoglou, Barbara Spagnolo, Antonio Balena, Massimo De Vittorio, Ferruccio Pisanello   +4 more
wiley   +1 more source

FedRAS: a dual-strategy framework for federated learning on heterogeneous devices

Journal of Cloud Computing: Advances, Systems and Applications
Federated learning enables multiple clients to collaboratively train a global model using their respective local datasets. While it offers advantages such as privacy preservation and efficient data utilization, its practical deployment is still ...
Mohan Xu, Lena Wiese
doaj   +1 more source