Kernel Regression based Sparse Metric Learning for Extensive Classification of Visual Art Images

2021 ◽  
Author(s):  
D N S Ravi Kumar ◽  
G T Sundarrajan ◽  
S D Sundarsingh Jebaseelan ◽  
M. Pushpavalli ◽  
A Rameshbabu ◽  
...  
Keyword(s):  
Visual Art ◽  
Metric Learning ◽  
Kernel Regression

Creating with Constraints

2020 ◽  
Author(s):  
Robert Hasegawa
Keyword(s):  
Visual Art ◽  
Creative Activity ◽  
Georges Perec ◽  
Gyorgy Ligeti ◽  
Raymond Queneau ◽  
Art Forms ◽  
Musical Creativity ◽  
Process Constraints

Musicians have long framed their creative activity within constraints, whether imposed externally or consciously chosen. As noted by Leonard Meyer, any style can be viewed as an ensemble of constraints, requiring the features of the artwork to conform with accepted norms. Such received stylistic constraints may be complemented by additional, voluntary limitations: for example, using only a limited palette of pitches or sounds, setting rules to govern repetition or transformation, controlling the formal layout and proportions of the work, or limiting the variety of operations involved in its creation. This chapter proposes a fourfold classification of the limits most often encountered in music creation into material (absolute and relative), formal, style/genre, and process constraints. The role of constraints as a spur and guide to musical creativity is explored in the domains of composition, improvisation, performance, and even listening, with examples drawn from contemporary composers including György Ligeti, George Aperghis, and James Tenney. Such musical constraints are comparable to self-imposed limitations in other art forms, from film (the Dogme 95 Manifesto) and visual art (Robert Morris’s Blind Time Drawings) to the writings of authors associated with the Oulipo (Ouvroir de littérature potentielle) such as Georges Perec and Raymond Queneau.

scholarly journals Automated classification of clinical trial eligibility criteria text based on ensemble learning and metric learning

2021 ◽  
Vol 21 (S2) ◽  
Author(s):  
Kun Zeng ◽  
Yibin Xu ◽  
Ge Lin ◽  
Likeng Liang ◽  
Tianyong Hao
Keyword(s):  
Clinical Trial ◽  
Ensemble Learning ◽  
Metric Learning ◽  
Classification Performance ◽  
Ensemble Model ◽  
Automated Classification ◽  
Eligibility Criteria ◽  
Data Imbalance ◽  
The Impact

Abstract Background Eligibility criteria are the primary strategy for screening the target participants of a clinical trial. Automated classification of clinical trial eligibility criteria text by using machine learning methods improves recruitment efficiency to reduce the cost of clinical research. However, existing methods suffer from poor classification performance due to the complexity and imbalance of eligibility criteria text data. Methods An ensemble learning-based model with metric learning is proposed for eligibility criteria classification. The model integrates a set of pre-trained models including Bidirectional Encoder Representations from Transformers (BERT), A Robustly Optimized BERT Pretraining Approach (RoBERTa), XLNet, Pre-training Text Encoders as Discriminators Rather Than Generators (ELECTRA), and Enhanced Representation through Knowledge Integration (ERNIE). Focal Loss is used as a loss function to address the data imbalance problem. Metric learning is employed to train the embedding of each base model for feature distinguish. Soft Voting is applied to achieve final classification of the ensemble model. The dataset is from the standard evaluation task 3 of 5th China Health Information Processing Conference containing 38,341 eligibility criteria text in 44 categories. Results Our ensemble method had an accuracy of 0.8497, a precision of 0.8229, and a recall of 0.8216 on the dataset. The macro F1-score was 0.8169, outperforming state-of-the-art baseline methods by 0.84% improvement on average. In addition, the performance improvement had a p-value of 2.152e-07 with a standard t-test, indicating that our model achieved a significant improvement. Conclusions A model for classifying eligibility criteria text of clinical trials based on multi-model ensemble learning and metric learning was proposed. The experiments demonstrated that the classification performance was improved by our ensemble model significantly. In addition, metric learning was able to improve word embedding representation and the focal loss reduced the impact of data imbalance to model performance.

scholarly journals Changes in the themes, aesthetics and functions of public images in Kyiv due to the influence of socio-cultural processes in the society of 1990-2010

2021 ◽  
pp. 183-194
Author(s):  
Mariia Ospishcheva-Pavlyshyn
Keyword(s):  
Contemporary Art ◽  
Visual Arts ◽  
Public Art ◽  
Visual Art ◽  
Rapid Development ◽  
Practical Aspect ◽  
Cultural Processes ◽  
The Impact ◽  
Theoretical Developments

On the back of the rapid development in public art in recent decades, and in particular graffiti and muralism, interest in them has grown significantly among cultural studies scholars, art critics, architects, sociologists, and urban planners. Numerous works that have appeared in the West and in Ukraine are devoted to various aspects of the visual public art existence. This theme continues to be one of the most relevant for contemporary visual art. This article complements the bunch of acquired knowledge with a detailed study of the impact of socio-cultural processes in society on the changes that took place in monumental painting, graffiti and muralism in Kyiv during 1990–2010, i.e. during the most important changes in politics and society in recent decades. The peculiarities of each historical stage of this influence are analysed and outlined in the study, and the theoretical analysis is displayed by the description of the most characteristic works. Most of them are researched in detail. In addition, the process of decline of monumental painting in the late 1980s and early 1990s is analysed, the factors of graffiti flourishing in the 1990s are identified and highlighted, and the origins of the rapid development of muralism after 2004 and especially after 2014 are explored. At each stage, changes in the themes, aesthetics and functions of public images are traced. The definitions, such as muralism and graffiti, are updated in this paper, taking into account changes in art and the latest achievements in its analysis. The manifestations of the national-patriotic themes in the contemporary art of muralism are considered in detail, the classification of art work on this subject is given, the corresponding examples are given. Such concepts as public art, synthesis of arts, monumental painting, graffiti, muralism are attentively aligned. The study of the nature of the socio-cultural processes and visual arts correlations is promising for further scientific and theoretical developments and the practical aspect for better understanding of the specific works

The Classification of Visual Art

2001 ◽  
Vol 41 (4) ◽  
pp. 457-458
Author(s):  
C. Lyas
Keyword(s):  
Visual Art

scholarly journals Semi-Supervised Deep Metric Learning Networks for Classification of Polarimetric SAR Data

2020 ◽  
Vol 12 (10) ◽  
pp. 1593
Author(s):  
Hongying Liu ◽  
Ruyi Luo ◽  
Fanhua Shang ◽  
Xuechun Meng ◽  
Shuiping Gou ◽  
...  
Keyword(s):  
Real World ◽  
Metric Learning ◽  
Feature Learning ◽  
Linear Mapping ◽  
Layer By Layer ◽  
Learning Networks ◽  
Distance Metric ◽  
Linear Projection ◽  
Deep Metric Learning

Recently, classification methods based on deep learning have attained sound results for the classification of Polarimetric synthetic aperture radar (PolSAR) data. However, they generally require a great deal of labeled data to train their models, which limits their potential real-world applications. This paper proposes a novel semi-supervised deep metric learning network (SSDMLN) for feature learning and classification of PolSAR data. Inspired by distance metric learning, we construct a network, which transforms the linear mapping of metric learning into the non-linear projection in the layer-by-layer learning. With the prior knowledge of the sample categories, the network also learns a distance metric under which all pairs of similarly labeled samples are closer and dissimilar samples have larger relative distances. Moreover, we introduce a new manifold regularization to reduce the distance between neighboring samples since they are more likely to be homogeneous. The categorizing is achieved by using a simple classifier. Several experiments on both synthetic and real-world PolSAR data from different sensors are conducted and they demonstrate the effectiveness of SSDMLN with limited labeled samples, and SSDMLN is superior to state-of-the-art methods.

Sign in / Sign up

Export Citation Format

Share Document