Results 81 to 90 of about 6,336,923 (296)
Avalanche: A PyTorch Library for Deep Continual Learning [PDF]
, 2023 Continual learning is the problem of learning from a nonstationary stream of data, a fundamental issue for sustainable and efficient training of deep neural networks over time.
Lorenzo Pellegrini +4 more
core +2 more sources
, 2021 This work focuses on building frameworks to strengthen the relation between human and machine learning. This is achieved by proposing a new category of algorithms and a new theory to formalize the perception and categorization of objects. For what concerns the algorithmic part, we developed a series of procedures to perform Interactive Continuous Open ...
openaire +1 more source
Advanced Science, EarlyView.Computational simulations of tumor evolution are increasingly used to infer the rules underlying cancer growth, with the goal of one day recommending tailored treatments. Here we show that the properties of lung cancer sequencing data are best replicated by a model which assumes that cells compete both to proliferate and survive. ABSTRACT Computational
Helena Coggan +5 more
wiley +1 more source
Enhancing Generative Class Incremental Learning Performance With a Model Forgetting Approach
IEEE Open Journal of Signal ProcessingThis study presents a novel approach to Generative Class Incremental Learning (GCIL) by introducing the forgetting mechanism, aimed at dynamically managing class information for better adaptation to streaming data.
Taro Togo +4 more
doaj +1 more source
CoRRRegularizing continual learning techniques is important for anticipating algorithmic behavior under new realizations of data. We introduce a new approach to continual learning by imposing the properties of a parabolic partial differential equation (PDE) to regularize the expected behavior of the loss over time. This class of parabolic PDEs has a number
Haoming Yang, Ali Hasan, Vahid Tarokh
openaire +3 more sources
Variance-Covariance Regularization Improves Continual Learning
IEEE AccessIn this work, we explore the benefits of Variance-Covariance Regularization in Continual Learning (CL). Neural networks suffer from abrupt performance loss when updated with additional data.
Piotr Hondra, Daniel Marczak, Kamil Deja
doaj +1 more source
Advanced Science, EarlyView.A clinically translatable cell line can be engineered to produce leptin, encapsulated in biomaterial and safely implanted to produce and deliver the metabolic regulating protein within the body. Implantation of these encapsulated cells decreases the time it takes for mice and non‐human primates to adjust to circadian disruptions similar to those ...
Samantha T. Fleury +18 more
wiley +1 more source
Scalable continual deep learning with computational cost considerations [PDF]
Continual Learning (CL) is a emerging field that focuses on developing models capable of learning continuously from a incoming stream of data, as opposed to hundreds of passes over static, curated datasets. These models aim to retain previously acquired
Prabhu, Ameya
core +1 more source
Advanced Science, EarlyView.An integrative review of microfluidics‐enabled wearables and implantable systems reveals a single‐track translation pipeline, bridging functional biomaterials with clinical utility. Dynamic feedback loops driven by artificial intelligence advance diagnostics toward personalized closed‐loop theranostics.
Ke Huang +3 more
wiley +1 more source
Embedded Continual Learning for High-Energy Physics [PDF]
EPJ Web of ConferencesNeural Networks (NN) are often trained offline on large datasets and deployed on specialised hardware for inference, with a strict separation between training and inference.
Barbone Marco +7 more
doaj +1 more source