32,162 Hits in 3.9 sec

Deep Bayesian Self-Training [article]

Fabio De Sousa Ribeiro, Francesco Caliva, Mark Swainson, Kjartan Gudmundsson, Georgios Leontidis, Stefanos Kollias
2019 arXiv   pre-print
Supervised Deep Learning has been highly successful in recent years, achieving state-of-the-art results in most tasks.  ...  In this paper, we propose both a (i) Deep Bayesian Self-Training methodology for automatic data annotation, by leveraging predictive uncertainty estimates using variational inference and modern Neural  ...  A large database, consisting of about half a million food packaging images has been obtained, and the intend to apply the presented deep neural network based methodologies for adaptation and self-annotation  ... 
arXiv:1812.01681v3 fatcat:xojudkc2mrarpjwnl3z4o5egq4

Effects of the Nonlinearity in Activation Functions on the Performance of Deep Learning Models [article]

Nalinda Kulathunga, Nishath Rajiv Ranasinghe, Daniel Vrinceanu, Zackary Kinsman, Lei Huang, Yunjiao Wang
2020 arXiv   pre-print
Furthermore, we found that the image classification models seem to perform well with L-ReLU in fully connected layers, especially when pre-trained models such as the VGG-16 are used for the transfer learning  ...  The nonlinearity of activation functions used in deep learning models are crucial for the success of predictive models.  ...  ACKNOWLEDGMENT This work is funded by the National Science Foundation (NSF), grant no: CNS-1831980 and HRD-1800406.  ... 
arXiv:2010.07359v1 fatcat:xq2wz6pcrzd5daqqoji3pqttfm

Research status and applications of nature-inspired algorithms for agri-food production

Yanbo Huang, USDA-ARS Crop Production Systems Research Unit, Stoneville, MS 38776, USA
2020 International Journal of Agricultural and Biological Engineering  
Now, the research and applications have entered the stage of deep learning with more layers and neurons that have complex connections to extract deep features of the target.  ...  In this paper, the developments of artificial neural networks and deep learning algorithms are presented and discussed in conjunction with their biological connections for agri-food applications.  ...  To tackle the issues to ensure global food security, it is necessary to develop and apply advanced technologies such as artificial intelligence (AI) and nature-inspired computing in agricultural and food  ... 
doi:10.25165/j.ijabe.20201304.5501 fatcat:cq2moh4ep5b5phtgr7f4tixbje

Food recognition and recipe analysis: integrating visual content, context and external knowledge [article]

Luis Herranz, Weiqing Min, Shuqiang Jiang
2018 arXiv   pre-print
We review how visual content, context and external knowledge can be integrated effectively into food-oriented applications, with special focus on recipe analysis and retrieval, food recommendation, and  ...  The central role of food in our individual and social life, combined with recent technological advances, has motivated a growing interest in applications that help to better monitor dietary habits as well  ...  Visual features extracted from food images can also be valuable signals in food analysis. For instance, Yang et al. [8] use food images to learn the food preferences of users. Recently, Min et al.  ... 
arXiv:1801.07239v1 fatcat:kbcpto5iznhkddvdklwxxbtehm

Learning to Compare Relation: Semantic Alignment for Few-Shot Learning [article]

Congqi Cao, Yanning Zhang
2022 arXiv   pre-print
The representation and metric for comparison are critical but challenging to learn due to the scarcity and wide variation of the samples in few-shot learning.  ...  We propose to add two key ingredients to existing few-shot learning frameworks for better feature and metric learning ability.  ...  In data analysis, it is essential to not only provide a good model but also an uncertainty estimate of the conclusions.  ... 
arXiv:2003.00210v2 fatcat:ukrrdupfvnhpzbhviuqqfxtf7m

Computational biology: deep learning

William Jones, Kaur Alasoo, Dmytro Fishman, Leopold Parts
2017 Emerging Topics in Life Sciences  
In pioneering early work, applying simple network architectures to abundant data already provided gains over traditional counterparts in functional genomics, image analysis, and medical diagnostics.  ...  Now, ideas for constructing and training networks and even off-the-shelf models have been adapted from the rapidly developing machine learning subfield to improve performance in a range of computational  ...  Deep learning automates the quantitative analysis of individual cells in live-cell imaging experiments [47] CNN Microscopy images Cell segmentations Able to segment both mammalian and bacterial  ... 
doi:10.1042/etls20160025 pmid:33525807 pmcid:PMC7289034 fatcat:qnw2yndsp5aqlnxxshtaipzctu

A Hierarchical deep model for food classification from photographs

2020 KSII Transactions on Internet and Information Systems  
We build a hierarchical structure composed of deep CNN to recognize and classify food from photographs.  ...  Recent progress of deep learning techniques accelerates the recognition of food in a great scale.  ...  Joutou and Yanai [2] applied a multiple kernel learning approach to merge various features such as color, texture and SIFT.  ... 
doi:10.3837/tiis.2020.04.016 fatcat:5lyy6nflmncohocwabghesfxxm

Multi-class Text Classification using BERT-based Active Learning [article]

Sumanth Prabhu and Moosa Mohamed and Hemant Misra
2021 arXiv   pre-print
In this paper, we explore Active Learning strategies to label transaction descriptions cost effectively while using BERT to train a transaction classification model.  ...  However, the product descriptions provided by the customers tend to be short, incoherent and code-mixed (Hindi-English) text which demands fine-tuning of such models with manually labelled data to achieve  ...  However, Bayesian Deep Learning based approaches (Gal and Ghahramani, 2016; Gal et al., 2017; Kirsch et al., 2019) were leveraged to demonstrate high performance in an active learning setting for image  ... 
arXiv:2104.14289v2 fatcat:6syugixzbbgb7hm4l3uet7peg4

Bayesian deep learning of affordances from RGB images [article]

Lorenzo Mur-Labadia, Ruben Martinez-Cantin
2021 arXiv   pre-print
Our Bayesian model is able to capture both the aleatoric uncertainty from the scene and the epistemic uncertainty associated with the model and previous learning process.  ...  In this paper, we present a Bayesian deep learning method to predict the affordances available in the environment directly from RGB images.  ...  In [32] , they compare epistemic and aleatoric uncertainty using convolutional neural networks (CNNs) applied to medical image segmentation problems at pixel and structure levels.  ... 
arXiv:2109.12845v1 fatcat:n7gqlxomina75msjhret3gak7i

Deep Learning and Earth Observation to Support the Sustainable Development Goals [article]

Claudio Persello, Jan Dirk Wegner, Ronny Hänsch, Devis Tuia, Pedram Ghamisi, Mila Koeva, Gustau Camps-Valls
2021 arXiv   pre-print
The synergistic combination of deep learning models and Earth observation promises significant advances to support the sustainable development goals (SDGs).  ...  This paper reviews current deep learning approaches for Earth observation data, along with their application towards monitoring and achieving the SDGs most impacted by the rapid development of deep learning  ...  RNNs have also been applied to HSI image analysis.  ... 
arXiv:2112.11367v1 fatcat:7eve5dr45vcublfqyzzrccuvxa

Deep Learning and Machine Vision for Food Processing: A Survey [article]

Lili Zhu, Petros Spachos, Erica Pensini, Konstantinos Plataniotis
2021 arXiv   pre-print
Image processing can take advantage of machine learning and deep learning models to effectively identify the type and quality of food.  ...  In this paper, we provide an overview on the traditional machine learning and deep learning methods, as well as the machine vision techniques that can be applied to the field of food processing.  ...  ., 2017) Deep Learning Methods Deep learning, also known as a deep neural network, is a learning method for building a deep architecture and generating a model by iterating functions in multiple layers  ... 
arXiv:2103.16106v1 fatcat:jr3pw7a6inf2tlpef3fk3p2xma

Automatic Prediction of Ischemia-Reperfusion Injury of Small Intestine Using Convolutional Neural Networks: A Pilot Study

Jie Hou, Runar Strand-Amundsen, Christian Tronstad, Jan Olav Høgetveit, Ørjan Grøttem Martinsen, Tor Inge Tønnessen
2021 Sensors  
To be able to assess to what extent we can trust our deep learning model decisions is critical in a clinical setting.  ...  We compared how different deep learning models perform for this task.  ...  Acknowledgments: The authors would like to thank Arnstein Arnesen at Sensocure AS, Rafael Palomar at Oslo University Hospital, and the medical staff at the Department of Emergencies and Critical Care at  ... 
doi:10.3390/s21196691 pmid:34641009 fatcat:pil5wc53irerrcmuem3kgx5yly

Wide-Slice Residual Networks for Food Recognition [article]

Niki Martinel, Gian Luca Foresti, Christian Micheloni
2016 arXiv   pre-print
Food diary applications represent a tantalizing market. Such applications, based on image food recognition, opened to new challenges for computer vision and pattern recognition algorithms.  ...  Specifically, inspired by the recent success of residual deep network, we exploit such a learning scheme and introduce a slice convolution block to capture the vertical food layers.  ...  Such information, coupled with a multiple kernel learning scheme applied on different visual features yield to the food image classification. Calories Estimation.  ... 
arXiv:1612.06543v1 fatcat:zkyrdg6t2rdqvdr5sl2ogriigm

Deep learning and machine vision for food processing: A survey

Lili Zhu, Petros Spachos, Erica Pensini, Konstantinos N. Plataniotis
2021 Current Research in Food Science  
Image processing can take advantage of machine learning and deep learning models to effectively identify the type and quality of food.  ...  In this paper, we provide an overview on the traditional machine learning and deep learning methods, as well as the machine vision techniques that can be applied to the field of food processing.  ...  Deep learning methods Deep learning, also known as a deep neural network, is a learning method for building a deep architecture and generating a model by iterating functions in multiple layers.  ... 
doi:10.1016/j.crfs.2021.03.009 pmid:33937871 pmcid:PMC8079277 fatcat:cqzvzbwwjrdulnve6shf7o2agu

Winter Wheat Yield Prediction at County Level and Uncertainty Analysis in Main Wheat-Producing Regions of China with Deep Learning Approaches

Xinlei Wang, Jianxi Huang, Quanlong Feng, Dongqin Yin
2020 Remote Sensing  
We further conducted yield prediction and uncertainty analysis based on the two-branch model and obtained the forecast accuracy in one month prior to harvest of 0.75 and 732 kg/ha.  ...  In this study, we established a two-branch deep learning model to predict winter wheat yield in the main producing regions of China at the county level.  ...  In recent years, machine learning and deep learning techniques have been successfully applied in many areas, such as image recognition, language translation, and signal processing [15] .  ... 
doi:10.3390/rs12111744 fatcat:tokajpou7rcptbup4ivqzvj6pm
« Previous Showing results 1 — 15 out of 32,162 results