scholarly journals Semisupervised Deep Embedded Clustering with Adaptive Labels

2021 ◽  
Vol 2021 ◽  
pp. 1-12
Author(s):  
Zhikui Chen ◽  
Chaojie Li ◽  
Jing Gao ◽  
Jianing Zhang ◽  
Peng Li
Keyword(s):  
Clustering Algorithm ◽  
State Of The Art ◽  
A Priori ◽  
Back Propagation ◽  
Normalized Mutual Information ◽  
Deep Embedding ◽  
End To End ◽  
Semisupervised Clustering ◽  
Fine Tune ◽  
Priori Knowledge

Deep embedding clustering (DEC) attracts much attention due to its outperforming performance attributed to the end-to-end clustering. However, DEC cannot make use of small amount of a priori knowledge contained in data of increasing volume. To tackle this challenge, a semisupervised deep embedded clustering algorithm with adaptive labels is proposed to cluster those data in a semisupervised end-to-end manner on the basis of a little priori knowledge. Specifically, a deep semisupervised clustering network is designed based on the autoencoder paradigm and deep clustering, which well mine the clustering representation and clustering assignment by preventing the shift of labels in DEC. Then, to train parameters of the deep semisupervised clustering network, a back-propagation-based algorithm with adaptive labels is introduced based on the pretrain and fine-tune strategies. Finally, extensive experiments on representative datasets are conducted to evaluate the performance of the proposed method in terms of clustering accuracy and normalized mutual information. Results show the proposed method outperforms the state-of-the-art methods of DEC.

Building Attention and Edge Convolution Neural Networks for Bioactivity and Physical-Chemical Property Prediction

2019 ◽  
Author(s):  
Michael Withnall ◽  
Edvard Lindelöf ◽  
Ola Engkvist ◽  
Hongming Chen
Keyword(s):  
Message Passing ◽  
State Of The Art ◽  
A Priori ◽  
Model Performance ◽  
Learning Approaches ◽  
Physical Chemical ◽  
Chemical Descriptor ◽  
Derived Properties ◽  
Memory Schemes ◽  
Hyperparameter Selection

We introduce Attention and Edge Memory schemes to the existing Message Passing Neural Network framework for graph convolution, and benchmark our approaches against eight different physical-chemical and bioactivity datasets from the literature. We remove the need to introduce <i>a priori</i> knowledge of the task and chemical descriptor calculation by using only fundamental graph-derived properties. Our results consistently perform on-par with other state-of-the-art machine learning approaches, and set a new standard on sparse multi-task virtual screening targets. We also investigate model performance as a function of dataset preprocessing, and make some suggestions regarding hyperparameter selection.

Segmentation of SAR Image using Fuzzy C-Means and Filters

2020 ◽  
Vol 8 (1) ◽  
pp. 84-90
Author(s):  
R. Lalchhanhima ◽  
◽  
Debdatta Kandar ◽  
R. Chawngsangpuii ◽  
Vanlalmuansangi Khenglawt ◽  
...  
Keyword(s):  
Clustering Algorithm ◽  
State Of The Art ◽  
Speckle Noise ◽  
Synthetic Aperture Radar Image ◽  
Synthetic Aperture ◽  
Sar Image ◽  
Spatial Filters ◽  
Fuzzy C Means ◽  
Automatic Clustering ◽  
Intensity Information

Fuzzy C-Means is an unsupervised clustering algorithm for the automatic clustering of data. Synthetic Aperture Radar Image Segmentation has been a challenging task because of the presence of speckle noise. Therefore the segmentation process can not directly rely on the intensity information alone but must consider several derived features in order to get satisfactory segmentation results. In this paper, it is attempted to use the fuzzy nature of classification for the purpose of unsupervised region segmentation in which FCM is employed. Different features are obtained by filtering of the image by using different spatial filters and are selected for segmentation criteria. The segmentation performance is determined by the accuracy compared with a different state of the art techniques proposed recently.

Seeing, Knowing, and Doing

2020 ◽  
Author(s):  
Robert Audi
Keyword(s):  
Practical Reasoning ◽  
Practical Knowledge ◽  
A Priori ◽  
Human Action ◽  
A Priori Knowledge ◽  
Rational Action ◽  
Reasons For Action ◽  
Discriminative Response ◽  
Rich Information ◽  
Priori Knowledge

This book provides an overall theory of perception and an account of knowledge and justification concerning the physical, the abstract, and the normative. It has the rigor appropriate for professionals but explains its main points using concrete examples. It accounts for two important aspects of perception on which philosophers have said too little: its relevance to a priori knowledge—traditionally conceived as independent of perception—and its role in human action. Overall, the book provides a full-scale account of perception, presents a theory of the a priori, and explains how perception guides action. It also clarifies the relation between action and practical reasoning; the notion of rational action; and the relation between propositional and practical knowledge. Part One develops a theory of perception as experiential, representational, and causally connected with its objects: as a discriminative response to those objects, embodying phenomenally distinctive elements; and as yielding rich information that underlies human knowledge. Part Two presents a theory of self-evidence and the a priori. The theory is perceptualist in explicating the apprehension of a priori truths by articulating its parallels to perception. The theory unifies empirical and a priori knowledge by clarifying their reliable connections with their objects—connections many have thought impossible for a priori knowledge as about the abstract. Part Three explores how perception guides action; the relation between knowing how and knowing that; the nature of reasons for action; the role of inference in determining action; and the overall conditions for rational action.

How Reality is Reasonable

2018 ◽  
Author(s):  
Donald C. Williams
Keyword(s):  
A Priori ◽  
A Priori Knowledge ◽  
Dimensional Manifold ◽  
The World ◽  
Being There ◽  
Ways Of Being ◽  
Fundamental Entity ◽  
Spatiotemporal Relations ◽  
Priori Knowledge

This chapter begins with a systematic presentation of the doctrine of actualism. According to actualism, all that exists is actual, determinate, and of one way of being. There are no possible objects, nor is there any indeterminacy in the world. In addition, there are no ways of being. It is proposed that actual entities stand in three fundamental relations: mereological, spatiotemporal, and resemblance relations. These relations govern the fundamental entities. Each fundamental entity stands in parthood relations, spatiotemporal relations, and resemblance relations to other entities. The resulting picture is one that represents the world as a four-dimensional manifold of actual ‘qualitied contents’—upon which all else supervenes. It is then explained how actualism accounts for classes, quantity, number, causation, laws, a priori knowledge, necessity, and induction.

How Do We Know that We’re Not Brains in Vats?

2018 ◽  
Author(s):  
Keith DeRose
Keyword(s):  
Epistemic Justification ◽  
A Priori ◽  
A Priori Knowledge ◽  
Conservative Approach ◽  
Contingent Fact ◽  
Brains In Vats ◽  
Priori Knowledge

In this chapter the contextualist Moorean account of how we know by ordinary standards that we are not brains in vats (BIVs) utilized in Chapter 1 is developed and defended, and the picture of knowledge and justification that emerges is explained. The account (a) is based on a double-safety picture of knowledge; (b) has it that our knowledge that we’re not BIVs is in an important way a priori; and (c) is knowledge that is easily obtained, without any need for fancy philosophical arguments to the effect that we’re not BIVs; and the account is one that (d) utilizes a conservative approach to epistemic justification. Special attention is devoted to defending the claim that we have a priori knowledge of the deeply contingent fact that we’re not BIVs, and to distinguishing this a prioritist account of this knowledge from the kind of “dogmatist” account prominently championed by James Pryor.

scholarly journals Mask-R$$^{2}$$CNN: a distance-field regression version of Mask-RCNN for fetal-head delineation in ultrasound images

2021 ◽  
Author(s):  
Sara Moccia ◽  
Maria Chiara Fiorentino ◽  
Emanuele Frontoni
Keyword(s):  
Fetal Growth ◽  
A Priori ◽  
Absolute Difference ◽  
Ultrasound Images ◽  
Fetal Head ◽  
Mean Absolute Difference ◽  
Distance Field ◽  
Feature Pyramid ◽  
End To End ◽  
Ablation Study

Abstract Background and objectives Fetal head-circumference (HC) measurement from ultrasound (US) images provides useful hints for assessing fetal growth. Such measurement is performed manually during the actual clinical practice, posing issues relevant to intra- and inter-clinician variability. This work presents a fully automatic, deep-learning-based approach to HC delineation, which we named Mask-R$$^{2}$$ 2 CNN. It advances our previous work in the field and performs HC distance-field regression in an end-to-end fashion, without requiring a priori HC localization nor any postprocessing for outlier removal. Methods Mask-R$$^{2}$$ 2 CNN follows the Mask-RCNN architecture, with a backbone inspired by feature-pyramid networks, a region-proposal network and the ROI align. The Mask-RCNN segmentation head is here modified to regress the HC distance field. Results Mask-R$$^{2}$$ 2 CNN was tested on the HC18 Challenge dataset, which consists of 999 training and 335 testing images. With a comprehensive ablation study, we showed that Mask-R$$^{2}$$ 2 CNN achieved a mean absolute difference of 1.95 mm (standard deviation $$=\pm 1.92$$ = ± 1.92  mm), outperforming other approaches in the literature. Conclusions With this work, we proposed an end-to-end model for HC distance-field regression. With our experimental results, we showed that Mask-R$$^{2}$$ 2 CNN may be an effective support for clinicians for assessing fetal growth.

scholarly journals Efficient End-to-End Sentence-Level Lipreading with Temporal Convolutional Networks

2021 ◽  
Vol 11 (15) ◽  
pp. 6975
Author(s):  
Tao Zhang ◽  
Lun He ◽  
Xudong Li ◽  
Guoqing Feng
Keyword(s):  
Performance Improvement ◽  
State Of The Art ◽  
Error Rates ◽  
Convolutional Network ◽  
Convolutional Networks ◽  
Sentence Level ◽  
End To End ◽  
High Level ◽  
Improved Accuracy ◽  
Talking Face

Lipreading aims to recognize sentences being spoken by a talking face. In recent years, the lipreading method has achieved a high level of accuracy on large datasets and made breakthrough progress. However, lipreading is still far from being solved, and existing methods tend to have high error rates on the wild data and have the defects of disappearing training gradient and slow convergence. To overcome these problems, we proposed an efficient end-to-end sentence-level lipreading model, using an encoder based on a 3D convolutional network, ResNet50, Temporal Convolutional Network (TCN), and a CTC objective function as the decoder. More importantly, the proposed architecture incorporates TCN as a feature learner to decode feature. It can partly eliminate the defects of RNN (LSTM, GRU) gradient disappearance and insufficient performance, and this yields notable performance improvement as well as faster convergence. Experiments show that the training and convergence speed are 50% faster than the state-of-the-art method, and improved accuracy by 2.4% on the GRID dataset.

scholarly journals Image Restoration by Learning Morphological Opening-Closing Network

2020 ◽  
Vol 4 (1) ◽  
pp. 87-107
Author(s):  
Ranjan Mondal ◽  
Moni Shankar Dey ◽  
Bhabatosh Chanda
Keyword(s):  
Neural Network ◽  
Image Restoration ◽  
State Of The Art ◽  
Source Code ◽  
Back Propagation ◽  
Image Features ◽  
Main Difficulty ◽  
The Right ◽  
Right Order ◽  
Morphological Opening

AbstractMathematical morphology is a powerful tool for image processing tasks. The main difficulty in designing mathematical morphological algorithm is deciding the order of operators/filters and the corresponding structuring elements (SEs). In this work, we develop morphological network composed of alternate sequences of dilation and erosion layers, which depending on learned SEs, may form opening or closing layers. These layers in the right order along with linear combination (of their outputs) are useful in extracting image features and processing them. Structuring elements in the network are learned by back-propagation method guided by minimization of the loss function. Efficacy of the proposed network is established by applying it to two interesting image restoration problems, namely de-raining and de-hazing. Results are comparable to that of many state-of-the-art algorithms for most of the images. It is also worth mentioning that the number of network parameters to handle is much less than that of popular convolutional neural network for similar tasks. The source code can be found here https://github.com/ranjanZ/Mophological-Opening-Closing-Net

Sign in / Sign up

Export Citation Format

Share Document