Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox

"Topic": models, code, and papers

Evaluating Adversarial Attacks on ImageNet: A Reality Check on Misclassification Classes

Nov 22, 2021
Utku Ozbulak, Maura Pintor, Arnout Van Messem, Wesley De Neve

Although ImageNet was initially proposed as a dataset for performance benchmarking in the domain of computer vision, it also enabled a variety of other research efforts. Adversarial machine learning is one such research effort, employing deceptive inputs to fool models in making wrong predictions. To evaluate attacks and defenses in the field of adversarial machine learning, ImageNet remains one of the most frequently used datasets. However, a topic that is yet to be investigated is the nature of the classes into which adversarial examples are misclassified. In this paper, we perform a detailed analysis of these misclassification classes, leveraging the ImageNet class hierarchy and measuring the relative positions of the aforementioned type of classes in the unperturbed origins of the adversarial examples. We find that $71\%$ of the adversarial examples that achieve model-to-model adversarial transferability are misclassified into one of the top-5 classes predicted for the underlying source images. We also find that a large subset of untargeted misclassifications are, in fact, misclassifications into semantically similar classes. Based on these findings, we discuss the need to take into account the ImageNet class hierarchy when evaluating untargeted adversarial successes. Furthermore, we advocate for future research efforts to incorporate categorical information.

* Accepted for publication in 35th Conference on Neural Information Processing Systems (NeurIPS 2021), Workshop on ImageNet: Past,Present, and Future 

  Access Paper or Ask Questions

Detecting Cross-Language Plagiarism using Open Knowledge Graphs

Nov 18, 2021
Johannes Stegm├╝ller, Fabian Bauer-Marquart, Norman Meuschke, Terry Ruas, Moritz Schubotz, Bela Gipp

Identifying cross-language plagiarism is challenging, especially for distant language pairs and sense-for-sense translations. We introduce the new multilingual retrieval model Cross-Language Ontology-Based Similarity Analysis (CL\nobreakdash-OSA) for this task. CL-OSA represents documents as entity vectors obtained from the open knowledge graph Wikidata. Opposed to other methods, CL\nobreakdash-OSA does not require computationally expensive machine translation, nor pre-training using comparable or parallel corpora. It reliably disambiguates homonyms and scales to allow its application to Web-scale document collections. We show that CL-OSA outperforms state-of-the-art methods for retrieving candidate documents from five large, topically diverse test corpora that include distant language pairs like Japanese-English. For identifying cross-language plagiarism at the character level, CL-OSA primarily improves the detection of sense-for-sense translations. For these challenging cases, CL-OSA's performance in terms of the well-established PlagDet score exceeds that of the best competitor by more than factor two. The code and data of our study are openly available.

* 10 pages, EEKE21, Preprint 

  Access Paper or Ask Questions

CCQA: A New Web-Scale Question Answering Dataset for Model Pre-Training

Oct 14, 2021
Patrick Huber, Armen Aghajanyan, Barlas O─čuz, Dmytro Okhonko, Wen-tau Yih, Sonal Gupta, Xilun Chen

With the rise of large-scale pre-trained language models, open-domain question-answering (ODQA) has become an important research topic in NLP. Based on the popular pre-training fine-tuning approach, we posit that an additional in-domain pre-training stage using a large-scale, natural, and diverse question-answering (QA) dataset can be beneficial for ODQA. Consequently, we propose a novel QA dataset based on the Common Crawl project in this paper. Using the readily available annotation, we extract around 130 million multilingual question-answer pairs, including about 60 million English data-points. With this previously unseen number of natural QA pairs, we pre-train popular language models to show the potential of large-scale in-domain pre-training for the task of question-answering. In our experiments, we find that pre-training question-answering models on our Common Crawl Question Answering dataset (CCQA) achieves promising results in zero-shot, low resource and fine-tuned settings across multiple tasks, models and benchmarks.

  Access Paper or Ask Questions

Hierarchical Relation-Guided Type-Sentence Alignment for Long-Tail Relation Extraction with Distant Supervision

Sep 19, 2021
Yang Li, Guodong Long, Tao Shen, Jing Jiang

Distant supervision uses triple facts in knowledge graphs to label a corpus for relation extraction, leading to wrong labeling and long-tail problems. Some works use the hierarchy of relations for knowledge transfer to long-tail relations. However, a coarse-grained relation often implies only an attribute (e.g., domain or topic) of the distant fact, making it hard to discriminate relations based solely on sentence semantics. One solution is resorting to entity types, but open questions remain about how to fully leverage the information of entity types and how to align multi-granular entity types with sentences. In this work, we propose a novel model to enrich distantly-supervised sentences with entity types. It consists of (1) a pairwise type-enriched sentence encoding module injecting both context-free and -related backgrounds to alleviate sentence-level wrong labeling, and (2) a hierarchical type-sentence alignment module enriching a sentence with the triple fact's basic attributes to support long-tail relations. Our model achieves new state-of-the-art results in overall and long-tail performance on benchmarks.

* 10 pages 

  Access Paper or Ask Questions

MAIR: Framework for mining relationships between research articles, strategies, and regulations in the field of explainable artificial intelligence

Jul 29, 2021
Stanisław Gizinski, Michał Kuzba, Bartosz Pielinski, Julian Sienkiewicz, Stanisław Łaniewski, Przemysław Biecek

The growing number of AI applications, also for high-stake decisions, increases the interest in Explainable and Interpretable Machine Learning (XI-ML). This trend can be seen both in the increasing number of regulations and strategies for developing trustworthy AI and the growing number of scientific papers dedicated to this topic. To ensure the sustainable development of AI, it is essential to understand the dynamics of the impact of regulation on research papers as well as the impact of scientific discourse on AI-related policies. This paper introduces a novel framework for joint analysis of AI-related policy documents and eXplainable Artificial Intelligence (XAI) research papers. The collected documents are enriched with metadata and interconnections, using various NLP methods combined with a methodology inspired by Institutional Grammar. Based on the information extracted from collected documents, we showcase a series of analyses that help understand interactions, similarities, and differences between documents at different stages of institutionalization. To the best of our knowledge, this is the first work to use automatic language analysis tools to understand the dynamics between XI-ML methods and regulations. We believe that such a system contributes to better cooperation between XAI researchers and AI policymakers.

  Access Paper or Ask Questions

MT: Multi-Perspective Feature Learning Network for Scene Text Detection

May 12, 2021
Chuang Yang, Mulin Chen, Yuan Yuan, Qi Wang

Text detection, the key technology for understanding scene text, has become an attractive research topic. For detecting various scene texts, researchers propose plenty of detectors with different advantages: detection-based models enjoy fast detection speed, and segmentation-based algorithms are not limited by text shapes. However, for most intelligent systems, the detector needs to detect arbitrary-shaped texts with high speed and accuracy simultaneously. Thus, in this study, we design an efficient pipeline named as MT, which can detect adhesive arbitrary-shaped texts with only a single binary mask in the inference stage. This paper presents the contributions on three aspects: (1) a light-weight detection framework is designed to speed up the inference process while keeping high detection accuracy; (2) a multi-perspective feature module is proposed to learn more discriminative representations to segment the mask accurately; (3) a multi-factor constraints IoU minimization loss is introduced for training the proposed model. The effectiveness of MT is evaluated on four real-world scene text datasets, and it surpasses all the state-of-the-art competitors to a large extent.

  Access Paper or Ask Questions

Comparative Study of Learning Outcomes for Online Learning Platforms

Apr 15, 2021
Francois St-Hilaire, Nathan Burns, Robert Belfer, Muhammad Shayan, Ariella Smofsky, Dung Do Vu, Antoine Frau, Joseph Potochny, Farid Faraji, Vincent Pavero, Neroli Ko, Ansona Onyi Ching, Sabina Elkins, Anush Stepanyan, Adela Matajova, Laurent Charlin, Yoshua Bengio, Iulian Vlad Serban, Ekaterina Kochmar

Personalization and active learning are key aspects to successful learning. These aspects are important to address in intelligent educational applications, as they help systems to adapt and close the gap between students with varying abilities, which becomes increasingly important in the context of online and distance learning. We run a comparative head-to-head study of learning outcomes for two popular online learning platforms: Platform A, which follows a traditional model delivering content over a series of lecture videos and multiple-choice quizzes, and Platform B, which creates a personalized learning environment and provides problem-solving exercises and personalized feedback. We report on the results of our study using pre- and post-assessment quizzes with participants taking courses on an introductory data science topic on two platforms. We observe a statistically significant increase in the learning outcomes on Platform B, highlighting the impact of well-designed and well-engineered technology supporting active learning and problem-based learning in online education. Moreover, the results of the self-assessment questionnaire, where participants reported on perceived learning gains, suggest that participants using Platform B improve their metacognition.

* 14 pages, 3 figures, 2 tables, accepted at AIED 2021 (2021 Conference on Artificial Intelligence in Education) 

  Access Paper or Ask Questions

Interleaving Learning, with Application to Neural Architecture Search

Mar 12, 2021
Hao Ban, Pengtao Xie

Interleaving learning is a human learning technique where a learner interleaves the studies of multiple topics, which increases long-term retention and improves ability to transfer learned knowledge. Inspired by the interleaving learning technique of humans, in this paper we explore whether this learning methodology is beneficial for improving the performance of machine learning models as well. We propose a novel machine learning framework referred to as interleaving learning (IL). In our framework, a set of models collaboratively learn a data encoder in an interleaving fashion: the encoder is trained by model 1 for a while, then passed to model 2 for further training, then model 3, and so on; after trained by all models, the encoder returns back to model 1 and is trained again, then moving to model 2, 3, etc. This process repeats for multiple rounds. Our framework is based on multi-level optimization consisting of multiple inter-connected learning stages. An efficient gradient-based algorithm is developed to solve the multi-level optimization problem. We apply interleaving learning to search neural architectures for image classification on CIFAR-10, CIFAR-100, and ImageNet. The effectiveness of our method is strongly demonstrated by the experimental results.

* arXiv admin note: substantial text overlap with arXiv:2012.04863 

  Access Paper or Ask Questions

Ferrofluidic Manipulator: Automatic Manipulation of Non-magnetic Microparticles at Air-Ferrofluid Interface

Dec 30, 2020
Zoran Cenev, P. A. Diluka Harischandra, Seppo Nurmi, Mika Latikka, Ville Hynninen, Robin H. A. Ras, Jaakko V. I. Timonen, Quan Zhou

Manipulation of small-scale matter is a fundamental topic in micro- and nanorobotics. Numerous magnetic robotic systems have been developed for the manipulation of microparticles in an ambient environment, liquid as well as on the air-liquid interface. Those systems move intrinsically magnetic or magnetically tagged objects by inducing a magnetic torque or force. However, most of the materials found in nature are non-magnetic. Here, we report a novel ferrofluidic manipulator for automatic two-dimensional manipulation of non-magnetic objects floating on top of a ferrofluid. The manipulation system employs eight centimeter-scale solenoids, which can move non-magnetic particles floating on the air-liquid interface by deforming the air-ferrofluid interface. Using linear programming, we can control the motion of non-magnetic particles with a predefined trajectory of a line, square, and circle with a precision of 57.4+/-33.6 um, 74+/-44.4 um, and 67.2+/-38.6 um, respectively. The ferrofluidic manipulator is versatile with the materials and the shapes of the objects under manipulation. We have successfully manipulated particles made of polyethylene, polystyrene, a silicon chip, and poppy and sesame seeds.

* 8 pages, 8 figures 

  Access Paper or Ask Questions

A Comprehensive Survey on Deep Music Generation: Multi-level Representations, Algorithms, Evaluations, and Future Directions

Nov 13, 2020
Shulei Ji, Jing Luo, Xinyu Yang

The utilization of deep learning techniques in generating various contents (such as image, text, etc.) has become a trend. Especially music, the topic of this paper, has attracted widespread attention of countless researchers.The whole process of producing music can be divided into three stages, corresponding to the three levels of music generation: score generation produces scores, performance generation adds performance characteristics to the scores, and audio generation converts scores with performance characteristics into audio by assigning timbre or generates music in audio format directly. Previous surveys have explored the network models employed in the field of automatic music generation. However, the development history, the model evolution, as well as the pros and cons of same music generation task have not been clearly illustrated. This paper attempts to provide an overview of various composition tasks under different music generation levels, covering most of the currently popular music generation tasks using deep learning. In addition, we summarize the datasets suitable for diverse tasks, discuss the music representations, the evaluation methods as well as the challenges under different levels, and finally point out several future directions.

* 96 pages,this is a draft 

  Access Paper or Ask Questions