scholarly journals Ordinal Zero-Shot Learning

2017 ◽  
Author(s):  
Zengwei Huo ◽  
Xin Geng
Keyword(s):  
Side Information ◽  
Training Data ◽  
Head Pose Estimation ◽  
Classification Problems ◽  
Ordinal Classification ◽  
Regression Methods ◽  
New Class ◽  
Text Corpora ◽  
Class Labels ◽  
Better Than

Zero-shot learning predicts new class even if no training data is available for that class. The solution to conventional zero-shot learning usually depends on side information such as attribute or text corpora. But these side information is not easy to obtain or use. Fortunately in many classification tasks, the class labels are ordered, and therefore closely related to each other. This paper deals with zero-shot learning for ordinal classification. The key idea is using label relevance to expand supervision information from seen labels to unseen labels. The proposed method SIDL generates a supervision intensity distribution (SID) that contains each label's supervision intensity, and then learns a mapping from instance to SID. Experiments on two typical ordinal classification problems, i.e., head pose estimation and age estimation, show that SIDL performs significantly better than the compared regression methods. Furthermore, SIDL appears much more robust against the increase of unseen labels than other compared baselines.

scholarly journals A Girl Has No Name: Automated Authorship Obfuscation using Mutant-X

2019 ◽  
Vol 2019 (4) ◽  
pp. 54-71
Author(s):  
Asad Mahmood ◽  
Faizan Ahmad ◽  
Zubair Shafiq ◽  
Padmini Srinivasan ◽  
Fareed Zaffar
Keyword(s):  
State Of The Art ◽  
Fitness Function ◽  
Random Search ◽  
Training Data ◽  
Authorship Attribution ◽  
Future Research ◽  
Original Text ◽  
Text Corpora ◽  
Semantic Relevance ◽  
Better Than

Abstract Stylometric authorship attribution aims to identify an anonymous or disputed document’s author by examining its writing style. The development of powerful machine learning based stylometric authorship attribution methods presents a serious privacy threat for individuals such as journalists and activists who wish to publish anonymously. Researchers have proposed several authorship obfuscation approaches that try to make appropriate changes (e.g. word/phrase replacements) to evade attribution while preserving semantics. Unfortunately, existing authorship obfuscation approaches are lacking because they either require some manual effort, require significant training data, or do not work for long documents. To address these limitations, we propose a genetic algorithm based random search framework called Mutant-X which can automatically obfuscate text to successfully evade attribution while keeping the semantics of the obfuscated text similar to the original text. Specifically, Mutant-X sequentially makes changes in the text using mutation and crossover techniques while being guided by a fitness function that takes into account both attribution probability and semantic relevance. While Mutant-X requires black-box knowledge of the adversary’s classifier, it does not require any additional training data and also works on documents of any length. We evaluate Mutant-X against a variety of authorship attribution methods on two different text corpora. Our results show that Mutant-X can decrease the accuracy of state-of-the-art authorship attribution methods by as much as 64% while preserving the semantics much better than existing automated authorship obfuscation approaches. While Mutant-X advances the state-of-the-art in automated authorship obfuscation, we find that it does not generalize to a stronger threat model where the adversary uses a different attribution classifier than what Mutant-X assumes. Our findings warrant the need for future research to improve the generalizability (or transferability) of automated authorship obfuscation approaches.

scholarly journals CAWA: An Attention-Network for Credit Attribution

2020 ◽  
Vol 34 (05) ◽  
pp. 8472-8479
Author(s):  
Saurav Manchanda ◽  
George Karypis
Keyword(s):  
State Of The Art ◽  
Text Summarization ◽  
Training Data ◽  
Learning Framework ◽  
Distant Supervision ◽  
Sentence Level ◽  
Class Labels ◽  
The Individual ◽  
Traditional Approaches ◽  
Better Than

Credit attribution is the task of associating individual parts in a document with their most appropriate class labels. It is an important task with applications to information retrieval and text summarization. When labeled training data is available, traditional approaches for sequence tagging can be used for credit attribution. However, generating such labeled datasets is expensive and time-consuming. In this paper, we present Credit Attribution With Attention (CAWA), a neural-network-based approach, that instead of using sentence-level labeled data, uses the set of class labels that are associated with an entire document as a source of distant-supervision. CAWA combines an attention mechanism with a multilabel classifier into an end-to-end learning framework to perform credit attribution. CAWA labels the individual sentences from the input document using the resultant attention-weights. CAWA improves upon the state-of-the-art credit attribution approach by not constraining a sentence to belong to just one class, but modeling each sentence as a distribution over all classes, leading to better modeling of semantically-similar classes. Experiments on the credit attribution task on a variety of datasets show that the sentence class labels generated by CAWA outperform the competing approaches. Additionally, on the multilabel text classification task, CAWA performs better than the competing credit attribution approaches1.

Semi-Supervised Learning

2011 ◽  
pp. 1787-1793
Author(s):  
Tobias Scheffer
Keyword(s):  
Active Learning ◽  
Supervised Learning ◽  
Error Rate ◽  
Learning Algorithms ◽  
Gaussian Mixture ◽  
Unlabeled Data ◽  
Training Data ◽  
Support Vector ◽  
Classification Problems ◽  
Class Labels

For many classification problems, unlabeled training data are inexpensive and readily available, whereas labeling training data imposes costs. Semi-supervised classification algorithms aim at utilizing information contained in unlabeled data in addition to the (few) labeled data. Semi-supervised (for an example, see Seeger, 2001) has a long tradition in statistics (Cooper & Freeman, 1970); much early work has focused on Bayesian discrimination of Gaussians. The Expectation Maximization (EM) algorithm (Dempster, Laird, & Rubin, 1977) is the most popular method for learning generative models from labeled and unlabeled data. Model-based, generative learning algorithms find model parameters (e.g., the parameters of a Gaussian mixture model) that best explain the available labeled and unlabeled data, and they derive the discriminating classification hypothesis from this model. In discriminative learning, unlabeled data is typically incorporated via the integration of some model assumption into the discriminative framework (Miller & Uyar, 1997; Titterington, Smith, & Makov, 1985). The Transductive Support Vector Machine (Vapnik, 1998; Joachims, 1999) uses unlabeled data to identify a hyperplane that has a large distance not only from the labeled data but also from all unlabeled data. This identification results in a bias toward placing the hyperplane in regions of low density p(x). Recently, studies have covered graph-based approaches that rely on the assumption that neighboring instances are more likely to belong to the same class than remote instances (Blum & Chawla, 2001). A distinct approach to utilizing unlabeled data has been proposed by de Sa (1994), Yarowsky (1995) and Blum and Mitchell (1998). When the available attributes can be split into independent and compatible subsets, then multi-view learning algorithms can be employed. Multi-view algorithms, such as co-training (Blum & Mitchell, 1998) and co-EM (Nigam & Ghani, 2000), learn two independent hypotheses, which bootstrap by providing each other with labels for the unlabeled data. An analysis of why training two independent hypotheses that provide each other with conjectured class labels for unlabeled data might be better than EM-like self-training has been provided by Dasgupta, Littman, and McAllester (2001) and has been simplified by Abney (2002). The disagreement rate of two independent hypotheses is an upper bound on the error rate of either hypothesis. Multi-view algorithms minimize the disagreement rate between the peer hypotheses (a situation that is most apparent for the algorithm of Collins & Singer, 1999) and thereby the error rate. Semi-supervised learning is related to active learning. Active learning algorithms are able to actively query the class labels of unlabeled data. By contrast, semi-supervised algorithms are bound to learn from the given data.

THE CONCEPTUALIZATION OF EMOTIONS AND AXIOLOGICAL CONCEPTS IN SLAVONIC PROVERBS AND IDIOMS: FROM CONSCIENCE TO ENVY

2019 ◽  
Vol 2019 (29) ◽  
pp. 248-271
Author(s):  
Oleh Tyshchenko
Keyword(s):  
Cultural Anthropology ◽  
Legal Aspect ◽  
English Text ◽  
Folk Culture ◽  
Cultural Concepts ◽  
Text Corpora ◽  
Psychic Disorders ◽  
Body And Soul ◽  
Germanic Languages ◽  
Better Than

The presented research reveals imagery-metaphoric and phraseological objectivities of the conceptual spheres Soul, Consciousness, Envy, Jealousy and Greed in Polish, Russian, Ukrainian, Czech and Slovak languages and conceptual picture of the world (first of all in proverbs and sayings, idioms, imagery means of secondary nomination both in standard language and its regional or dialectal variants) according to the indication of holistic characteristic and semantic intersection of these concepts. It describes the spheres of their typological coincidence and differences from the point of imagery motivation. It defines the symbolic functions of these ethno cultural concepts (object sphere) with respect to the specificity of manifestation of Envy in archaic texts, believes, in the language of traditional folk culture and archaic expressions with religious sense that reach Christian ideology, ideas of moral purity and dirt, Body and Soul. It has been defined the collocations with the components envy and jealousy in some thesauri and dictionaries in terms of the specificity of interlingual equivalence and expressions of envy and similar negative emotions and their functioning in the Ukrainian and English text corpora. The analysis demonstrated that practically in all compared languages and linguistic cultures Envy is associated with greed and jealousy, psychic disorders with a corresponding complex of feelings, expressed by metaphoric predicates of destruction and remorse that encode the moral and legal aspect of conscience (conscience is a judge, witness and executioner). Metaphor of Envy containing nominations of colours differ in the Slavonic and Germanic languages whereas those denoting spatial, gustatory, odour, acoustic and parametrical meaning are similar. Many imagery contexts of Envy correlate with such conceptual oppositions as richness and poverty, light and darkness; success is associated with the frames “foreign is better than domestic” where Envy encodes the meaning of encroachment upon another's property, “envy is better than sympathy”, “envy dominates where there are richness, success, welfare, happiness” which confirms the ideas of representatives in the field of psychoanalysis, cultural anthropology and sociology. In some languages the motives of black magic, evil eye (in Polish, Ukrainian and Russian) are rooted in the sphere of folk believes and invocations, as well as cultural anthroponyms.

scholarly journals An analytics-based heuristic decomposition of a bilevel multiple-follower cutting stock problem

OR Spectrum ◽  
2021 ◽  
Author(s):  
Adejuyigbe O. Fajemisin ◽  
Laura Climent ◽  
Steven D. Prestwich
Keyword(s):  
Real Life ◽  
Forest Harvesting ◽  
Cutting Stock Problem ◽  
Cutting Stock ◽  
Decomposition Approach ◽  
New Class ◽  
Bilevel Problem ◽  
Programming Techniques ◽  
Bilevel Problems ◽  
Better Than

AbstractThis paper presents a new class of multiple-follower bilevel problems and a heuristic approach to solving them. In this new class of problems, the followers may be nonlinear, do not share constraints or variables, and are at most weakly constrained. This allows the leader variables to be partitioned among the followers. We show that current approaches for solving multiple-follower problems are unsuitable for our new class of problems and instead we propose a novel analytics-based heuristic decomposition approach. This approach uses Monte Carlo simulation and k-medoids clustering to reduce the bilevel problem to a single level, which can then be solved using integer programming techniques. The examples presented show that our approach produces better solutions and scales up better than the other approaches in the literature. Furthermore, for large problems, we combine our approach with the use of self-organising maps in place of k-medoids clustering, which significantly reduces the clustering times. Finally, we apply our approach to a real-life cutting stock problem. Here a forest harvesting problem is reformulated as a multiple-follower bilevel problem and solved using our approach.

scholarly journals NLOS Multipath Classification of GNSS Signal Correlation Output Using Machine Learning

Sensors ◽  
2021 ◽  
Vol 21 (7) ◽  
pp. 2503
Author(s):  
Taro Suzuki ◽  
Yoshiharu Amano
Keyword(s):  
Machine Learning ◽  
Satellite System ◽  
Training Data ◽  
Support Vector ◽  
Positioning Errors ◽  
Automated Method ◽  
Global Navigation Satellite ◽  
Better Than ◽  
Signal Correlation

This paper proposes a method for detecting non-line-of-sight (NLOS) multipath, which causes large positioning errors in a global navigation satellite system (GNSS). We use GNSS signal correlation output, which is the most primitive GNSS signal processing output, to detect NLOS multipath based on machine learning. The shape of the multi-correlator outputs is distorted due to the NLOS multipath. The features of the shape of the multi-correlator are used to discriminate the NLOS multipath. We implement two supervised learning methods, a support vector machine (SVM) and a neural network (NN), and compare their performance. In addition, we also propose an automated method of collecting training data for LOS and NLOS signals of machine learning. The evaluation of the proposed NLOS detection method in an urban environment confirmed that NN was better than SVM, and 97.7% of NLOS signals were correctly discriminated.

scholarly journals Tag N’ Train: a technique to train improved classifiers on unlabeled data

2021 ◽  
Vol 2021 (1) ◽  
Author(s):  
Oz Amram ◽  
Cristina Mantilla Suarez
Keyword(s):  
Real Data ◽  
Unlabeled Data ◽  
Machine Learning Techniques ◽  
Jet Physics ◽  
Classification Problems ◽  
Weak Classifier ◽  
Potential Applications ◽  
Substantial Progress ◽  
Resonance Search ◽  
Class Labels

Abstract There has been substantial progress in applying machine learning techniques to classification problems in collider and jet physics. But as these techniques grow in sophistication, they are becoming more sensitive to subtle features of jets that may not be well modeled in simulation. Therefore, relying on simulations for training will lead to sub-optimal performance in data, but the lack of true class labels makes it difficult to train on real data. To address this challenge we introduce a new approach, called Tag N’ Train (TNT), that can be applied to unlabeled data that has two distinct sub-objects. The technique uses a weak classifier for one of the objects to tag signal-rich and background-rich samples. These samples are then used to train a stronger classifier for the other object. We demonstrate the power of this method by applying it to a dijet resonance search. By starting with autoencoders trained directly on data as the weak classifiers, we use TNT to train substantially improved classifiers. We show that Tag N’ Train can be a powerful tool in model-agnostic searches and discuss other potential applications.

scholarly journals Age Prediction of Human Based on DNA Methylation by Blood Tissues

Genes ◽  
2021 ◽  
Vol 12 (6) ◽  
pp. 870
Author(s):  
Jiansheng Zhang ◽  
Hongli Fu ◽  
Yan Xu
Keyword(s):  
Dna Methylation ◽  
Pearson Correlation ◽  
Close Correlation ◽  
Gradient Boosting ◽  
Tissue Samples ◽  
Regression Methods ◽  
Cpg Sites ◽  
Testing Dataset ◽  
Age Related ◽  
Better Than

In recent years, scientists have found a close correlation between DNA methylation and aging in epigenetics. With the in-depth research in the field of DNA methylation, researchers have established a quantitative statistical relationship to predict the individual ages. This work used human blood tissue samples to study the association between age and DNA methylation. We built two predictors based on healthy and disease data, respectively. For the health data, we retrieved a total of 1191 samples from four previous reports. By calculating the Pearson correlation coefficient between age and DNA methylation values, 111 age-related CpG sites were selected. Gradient boosting regression was utilized to build the predictive model and obtained the R2 value of 0.86 and MAD of 3.90 years on testing dataset, which were better than other four regression methods as well as Horvath’s results. For the disease data, 354 rheumatoid arthritis samples were retrieved from a previous study. Then, 45 CpG sites were selected to build the predictor and the corresponded MAD and R2 were 3.11 years and 0.89 on the testing dataset respectively, which showed the robustness of our predictor. Our results were better than the ones from other four regression methods. Finally, we also analyzed the twenty-four common CpG sites in both healthy and disease datasets which illustrated the functional relevance of the selected CpG sites.

Synthesis of trans-2-[N-(2-Hydroxy-1,2,3,4-tetrahydronaphthalene-1-yl)]iminothiazolidine and Related Compounds - A New Class of Antidepressants

1992 ◽  
Vol 57 (2) ◽  
pp. 415-424 ◽  
Author(s):  
Upendra K. Shukla ◽  
Raieshwar Singh ◽  
J. M. Khanna ◽  
Anil K. Saxena ◽  
Hemant K. Singh ◽  
...  
Keyword(s):  
Antidepressant Activity ◽  
Antiparasitic Activity ◽  
New Class ◽  
Related Compounds ◽  
Better Than

Antiparasitic and antidepressant activities exhibited by tetramisole (I) and its enantiomers prompted the study of its structural analogs trans-2-[N-(2-hydroxy-1,2,3,4-tetrahydronaphthalene/indane-1-yl)]iminothiazolidine (VIII/IX) and 2,3,4a,5,6,10b-hexahydronaphtho[1',2':4,5]-imidazo[2,1-b]thiazole (XII), 2,3,4a,5-tetrahydro-9bH-indeno[1',2':4,5]imidazo[2,1-b]thiazole (XIII), and 2,3,4a,5-tetrahydro-9bH-indeno[1',2':4,5]imidazo[2,1-b]thiazole (XVI), and a homolog 3,4,6,7-tetrahydro-7-phenyl-2H-imidazo[2,1-b]-1,3-thiazine (XX). While none of these compounds showed any noteworthy antiparasitic activity, the trans-2-[N-(2-hydroxy-1,2,3,4-tetrahydronaphthalene-1-yl)]iminothiazolidine (VIII) has shown marked antidepressant activity, better than imipramine in the tests used, and provides a new structural lead for antidepressants.

scholarly journals DeepIsoFun: a deep domain adaptation approach to predict isoform functions

2018 ◽  
Vol 35 (15) ◽  
pp. 2535-2544 ◽  
Author(s):  
Dipan Shaw ◽  
Hao Chen ◽  
Tao Jiang
Keyword(s):  
Network Architecture ◽  
Domain Adaptation ◽  
Multiple Instance Learning ◽  
Training Data ◽  
Supplementary Information ◽  
Operating Characteristics ◽  
Neural Network Architecture ◽  
Receiver Operating Characteristics Curve ◽  
Gene Functions ◽  
Better Than

AbstractMotivationIsoforms are mRNAs produced from the same gene locus by alternative splicing and may have different functions. Although gene functions have been studied extensively, little is known about the specific functions of isoforms. Recently, some computational approaches based on multiple instance learning have been proposed to predict isoform functions from annotated gene functions and expression data, but their performance is far from being desirable primarily due to the lack of labeled training data. To improve the performance on this problem, we propose a novel deep learning method, DeepIsoFun, that combines multiple instance learning with domain adaptation. The latter technique helps to transfer the knowledge of gene functions to the prediction of isoform functions and provides additional labeled training data. Our model is trained on a deep neural network architecture so that it can adapt to different expression distributions associated with different gene ontology terms.ResultsWe evaluated the performance of DeepIsoFun on three expression datasets of human and mouse collected from SRA studies at different times. On each dataset, DeepIsoFun performed significantly better than the existing methods. In terms of area under the receiver operating characteristics curve, our method acquired at least 26% improvement and in terms of area under the precision-recall curve, it acquired at least 10% improvement over the state-of-the-art methods. In addition, we also study the divergence of the functions predicted by our method for isoforms from the same gene and the overall correlation between expression similarity and the similarity of predicted functions.Availability and implementationhttps://github.com/dls03/DeepIsoFun/Supplementary informationSupplementary data are available at Bioinformatics online.

Sign in / Sign up

Export Citation Format

Share Document