Deep learning has achieved remarkable success in numerous domains with help from large amounts of big data. However, the quality of data labels is a concern because of the lack of high-quality labels ...in many real-world scenarios. As noisy labels severely degrade the generalization performance of deep neural networks, learning from noisy labels (robust training) is becoming an important task in modern deep learning applications. In this survey, we first describe the problem of learning with label noise from a supervised learning perspective. Next, we provide a comprehensive review of 62 state-of-the-art robust training methods, all of which are categorized into five groups according to their methodological difference, followed by a systematic comparison of six properties used to evaluate their superiority. Subsequently, we perform an in-depth analysis of noise rate estimation and summarize the typically used evaluation methodology, including public noisy datasets and evaluation metrics. Finally, we present several promising research directions that can serve as a guideline for future studies.
IBM POWER8 processor core microarchitecture Sinharoy, B.; Van Norstrand, J. A.; Eickemeyer, R. J. ...
IBM journal of research and development,
1/2015, Letnik:
59, Številka:
1
Journal Article
Object detection is a crucial task in computer vision systems with a wide range of applications in autonomous driving, medical imaging, retail, security, face recognition, robotics, and others. ...Nowadays, neural networks-based models are used to localize and classify instances of objects of particular classes. When real-time inference is not required, ensembles of models help to achieve better results.
In this work, we present a novel method for fusing predictions from different object detection models: weighted boxes fusion. Our algorithm utilizes confidence scores of all proposed bounding boxes to construct averaged boxes.
We tested the method on several datasets and evaluated it in the context of Open Images and COCO Object Detection challenges, achieving top results in these challenges. The 3D version of boxes fusion was successfully applied by the winning teams of Waymo Open Dataset and Lyft 3D Object Detection for Autonomous Vehicles challenges. The source code is publicly available at GitHub (Solovyev, 2019 31).
We present a novel method for combining predictions in ensembles of different object detection models: weighted boxes fusion. This method significantly improves the quality of the fused predicted rectangles for an ensemble.
We tested the method on several datasets and evaluated it in the context of the Open Images and COCO Object Detection challenges. It helped to achieve top results in these challenges. The source code is publicly available at GitHub.
•Novel method was proposed for combining predictions in ensembles of different object detection models.•Method significantly improves the quality of the fused predicted rectangles for an ensemble. The code is available at GitHub.•Method was tested on several datasets and evaluated in the context of the Open Images and COCO Object Detection challenges.
While improving prediction accuracy has been the focus of machine learning in recent years, this alone does not suffice for reliable decision-making. Deploying learning systems in consequential ...settings also requires calibrating and communicating the uncertainty of predictions. To convey instance-wise uncertainty for prediction tasks, we show how to generate set-valued predictions from a black-box predictor that controls the expected loss on future test points at a user-specified level. Our approach provides explicit finite-sample guarantees for any dataset by using a holdout set to calibrate the size of the prediction sets. This framework enables simple, distribution-free, rigorous error control for many tasks, and we demonstrate it in five large-scale machine learning problems: (1) classification problems where some mistakes are more costly than others; (2) multi-label classification, where each observation has multiple associated labels; (3) classification problems where the labels have a hierarchical structure; (4) image segmentation, where we wish to predict a set of pixels containing an object of interest; and (5) protein structure prediction. Last, we discuss extensions to uncertainty quantification for ranking, metric learning, and distributionally robust learning.
As a promising technology, edge computing extends computation, communication, and storage facilities toward the edge of a network. This new computing paradigm opens up new challenges, among which ...computation offloading is considered to be the most important one. Computation offloading enables end devices to offload computation tasks to edge servers and receive the results after the servers' execution of the tasks. In computation offloading, offloading modeling plays a crucial role in determining the overall edge computing performance. We present a comprehensive overview on the past development as well as the recent advances in research areas related to offloading modeling in edge computing. First, we present some important edge computing architectures and classify the previous works on computation offloading into different categories. Second, we discuss some basic models such as channel model, computation and communication model, and energy harvesting model that have been proposed in offloading modeling. Next, we elaborate on different offloading modeling methods which are based on (non-)convex optimization, Markov decision process, game theory, Lyapunov optimization, or machine learning. Finally, we highlight and discuss some research directions and challenges in the area of offloading modeling in edge computing.
Many natural combinatorial problems can be expressed as constraint satisfaction problems. This class of problems is known to be NP-complete in general, but certain restrictions on the form of the ...constraints can ensure tractability. The standard way to parameterize interesting subclasses of the constraint satisfaction problem is via finite constraint languages. The main problem is to classify those subclasses that are solvable in polynomial time and those that are NP-complete. It was conjectured that if a constraint language has a weak near-unanimity polymorphism then the corresponding constraint satisfaction problem is tractable; otherwise, it is NP-complete.
In the article, we present an algorithm that solves Constraint Satisfaction Problem in polynomial time for constraint languages having a weak near unanimity polymorphism, which proves the remaining part of the conjecture.
1
Label noise is an important issue in classification, with many potential negative consequences. For example, the accuracy of predictions may decrease, whereas the complexity of inferred models and ...the number of necessary training samples may increase. Many works in the literature have been devoted to the study of label noise and the development of techniques to deal with label noise. However, the field lacks a comprehensive survey on the different types of label noise, their consequences and the algorithms that consider label noise. This paper proposes to fill this gap. First, the definitions and sources of label noise are considered and a taxonomy of the types of label noise is proposed. Second, the potential consequences of label noise are discussed. Third, label noise-robust, label noise cleansing, and label noise-tolerant algorithms are reviewed. For each category of approaches, a short discussion is proposed to help the practitioner to choose the most suitable technique in its own particular field of application. Eventually, the design of experiments is also discussed, what may interest the researchers who would like to test their own algorithms. In this paper, label noise consists of mislabeled instances: no additional information is assumed to be available like e.g., confidences on labels.
Internet of Things (IoT) is a new paradigm that integrates the Internet and physical objects belonging to different domains such as home automation, industrial process, human health and environmental ...monitoring. It deepens the presence of Internet-connected devices in our daily activities, bringing, in addition to many benefits, challenges related to security issues. For more than two decades, Intrusion Detection Systems (IDS) have been an important tool for the protection of networks and information systems. However, applying traditional IDS techniques to IoT is difficult due to its particular characteristics such as constrained-resource devices, specific protocol stacks, and standards. In this paper, we present a survey of IDS research efforts for IoT. Our objective is to identify leading trends, open issues, and future research possibilities. We classified the IDSs proposed in the literature according to the following attributes: detection method, IDS placement strategy, security threat and validation strategy. We also discussed the different possibilities for each attribute, detailing aspects of works that either propose specific IDS schemes for IoT or develop attack detection strategies for IoT threats that might be embedded in IDSs.