Results 111 to 120 of about 2,291,078 (312)

Power Consumption Variation over Activation Functions [PDF]

arXiv, 2020
The power that machine learning models consume when making predictions can be affected by a model's architecture. This paper presents various estimates of power consumption for a range of different activation functions, a core factor in neural network model architecture design.
arxiv  

Addressing persistent challenges in digital image analysis of cancer tissue: resources developed from a hackathon

Molecular Oncology, EarlyView.
Large multidimensional digital images of cancer tissue are becoming prolific, but many challenges exist to automatically extract relevant information from them using computational tools. We describe publicly available resources that have been developed jointly by expert and non‐expert computational biologists working together during a virtual hackathon
Sandhya Prabhakaran, Clarence Yapp, Gregory J. Baker, Johanna Beyer, Young Hwan Chang, Allison L. Creason, Robert Krueger, Jeremy Muhlich, Nathan Heath Patterson, Kevin Sidak, Damir Sudar, Adam J. Taylor, Luke Ternes, Jakob Troidl, Xie Yubin, Artem Sokolov, Darren R. Tyson   +16 more
wiley   +1 more source

The Active Segmentation Platform for Microscopic Image Classification and Segmentation

Brain Sciences, 2021
Image segmentation still represents an active area of research since no universal solution can be identified. Traditional image segmentation algorithms are problem-specific and limited in scope.
Sumit K. Vohra, Dimiter Prodanov
doaj   +1 more source

Model Uncertainty based Active Learning on Tabular Data using Boosted Trees [PDF]

arXiv, 2023
Supervised machine learning relies on the availability of good labelled data for model training. Labelled data is acquired by human annotation, which is a cumbersome and costly process, often requiring subject matter experts. Active learning is a sub-field of machine learning which helps in obtaining the labelled data efficiently by selecting the most ...
arxiv  

Enzyme activity from machine learning [PDF]

Science, 2019
Enzyme Engineering Enzymes are very efficient catalysts for biochemical reactions, which are increasingly important for industrial applications. However, incomplete knowledge of the key factors that induce their catalytic properties limits our ability to engineer new enzymes with new properties. Bonk et al.
openaire   +2 more sources

Integration of single‐cell and bulk RNA‐sequencing data reveals the prognostic potential of epithelial gene markers for prostate cancer

Molecular Oncology, EarlyView.
Prostate cancer is a leading malignancy with significant clinical heterogeneity in men. An 11‐gene signature derived from dysregulated epithelial cell markers effectively predicted biochemical recurrence‐free survival in patients who underwent radical surgery or radiotherapy.
Zhuofan Mou, Lorna W. Harries
wiley   +1 more source

Using human brain activity to guide machine learning [PDF]

Scientific Reports, 2018
AbstractMachine learning is a field of computer science that builds algorithms that learn. In many cases, machine learning algorithms are used to recreate a human ability like adding a caption to a photo, driving a car, or playing a game. While the human brain has long served as a source ofinspirationfor machine learning, little effort has been made to
Fong, Ruth C., Scheirer, Walter, Cox, David   +2 more
openaire   +6 more sources

Exploration of heterogeneity and recurrence signatures in hepatocellular carcinoma

Molecular Oncology, EarlyView.
This study leveraged public datasets and integrative bioinformatic analysis to dissect malignant cell heterogeneity between relapsed and primary HCC, focusing on intercellular communication, differentiation status, metabolic activity, and transcriptomic profiles.
Wen‐Jing Wu, Jianchao Wang, Fuqing Chen, Xuefeng Wang, Bin Lan, Ruyi Fu, Hong Wen, Fangfang Chen, Wei Hong, Tian‐Yu Tang, Ying He, Gang Chen, Jianyin Zhou, Hai‐Long Piao, Di Chen, Shu‐Yong Lin   +15 more
wiley   +1 more source

A large‐scale retrospective study in metastatic breast cancer patients using circulating tumour DNA and machine learning to predict treatment outcome and progression‐free survival

Molecular Oncology, EarlyView.
There is an unmet need in metastatic breast cancer patients to monitor therapy response in real time. In this study, we show how a noninvasive and affordable strategy based on sequencing of plasma samples with longitudinal tracking of tumour fraction paired with a statistical model provides valuable information on treatment response in advance of the ...
Emma J. Beddowes, Mario Ortega Duran, Solon Karapanagiotis, Alistair Martin, Meiling Gao, Riccardo Masina, Ramona Woitek, James Tanner, Fleur Tippin, Justine Kane, Jonathan Lay, Anja Brouwer, Stephen‐John Sammut, Suet‐Feung Chin, Davina Gale, Dana W. Y. Tsui, Sarah‐Jane Dawson, Nitzan Rosenfeld, Maurizio Callari, Oscar M. Rueda, Carlos Caldas   +20 more
wiley   +1 more source

Deep Bayesian Active Learning with Image Data [PDF]

arXiv, 2017
Even though active learning forms an important pillar of machine learning, deep learning tools are not prevalent within it. Deep learning poses several difficulties when used in an active learning setting. First, active learning (AL) methods generally rely on being able to learn and update models from small amounts of data.
arxiv