Anomaly detection algorithms find extensive use in various fields. This area of research has recently made great advances thanks to deep learning. A recent method, the deep Support Vector Data Description (deep SVDD), which is inspired by the classic kernel-based Support Vector Data Description (SVDD), is capable of simultaneously learning a feature representation of the data and a data-enclosing hypersphere. The method has shown promising results in both unsupervised and semi-supervised settings. However, deep SVDD suffers from hypersphere collapse---also known as mode collapse---, if the architecture of the model does not comply with certain architectural constraints, e.g. the removal of bias terms. These constraints limit the adaptability of the model and in some cases, may affect the model performance due to learning sub-optimal features. In this work, we consider two regularizers to prevent hypersphere collapse in deep SVDD. The first regularizer is based on injecting random noise via the standard cross-entropy loss. The second regularizer penalizes the minibatch variance when it becomes too small. Moreover, we introduce an adaptive weighting scheme to control the amount of penalization between the SVDD loss and the respective regularizer. Our proposed regularized variants of deep SVDD show encouraging results and outperform a prominent state-of-the-art method on a setup where the anomalies have no apparent geometrical structure.

Recent advancements in audio event classification often ignore the structure and relation between the label classes available as prior information. This structure can be defined by ontology and augmented in the classifier as a form of domain knowledge. To capture such dependencies between the labels, we propose an ontology-aware neural network containing two components: feed-forward ontology layers and graph convolutional networks (GCN). The feed-forward ontology layers capture the intra-dependencies of labels between different levels of ontology. On the other hand, GCN mainly models inter-dependency structure of labels within an ontology level. The framework is evaluated on two benchmark datasets for single-label and multi-label audio event classification tasks. The results demonstrate the proposed solutions efficacy to capture and explore the ontology relations and improve the classification performance.

Learning Management Systems (LMS) and Educational Data Mining (EDM) are two important parts of online educational environment with the former being a centralised web-based information systems where the learning content is managed and learning activities are organised (Stone and Zheng,2014) and latter focusing on using data mining techniques for the analysis of data so generated. As part of this work, we present a literature review of three major tasks of EDM (See section 2), by identifying shortcomings and existing open problems, and a Blumenfield chart (See section 3). The consolidated set of papers and resources so used are released in https://github.com/manikandan-ravikiran/cs6460-Survey. The coverage statistics and review matrix of the survey are as shown in Figure 1 & Table 1 respectively. Acronym expansions are added in the Appendix Section 4.1.

Deep Reinforcement Learning (DRL) has numerous applications in the real world thanks to its outstanding ability in quickly adapting to the surrounding environments. Despite its great advantages, DRL is susceptible to adversarial attacks, which precludes its use in real-life critical systems and applications (e.g., smart grids, traffic controls, and autonomous vehicles) unless its vulnerabilities are addressed and mitigated. Thus, this paper provides a comprehensive survey that discusses emerging attacks in DRL-based systems and the potential countermeasures to defend against these attacks. We first cover some fundamental backgrounds about DRL and present emerging adversarial attacks on machine learning techniques. We then investigate more details of the vulnerabilities that the adversary can exploit to attack DRL along with the state-of-the-art countermeasures to prevent such attacks. Finally, we highlight open issues and research challenges for developing solutions to deal with attacks for DRL-based intelligent systems.

Millions of news articles are published online every day, which can be overwhelming for readers to follow. Grouping articles that are reporting the same event into news stories is a common way of assisting readers in their news consumption. However, it remains a challenging research problem to efficiently and effectively generate a representative headline for each story. Automatic summarization of a document set has been studied for decades, while few studies have focused on generating representative headlines for a set of articles. Unlike summaries, which aim to capture most information with least redundancy, headlines aim to capture information jointly shared by the story articles in short length, and exclude information that is too specific to each individual article. In this work, we study the problem of generating representative headlines for news stories. We develop a distant supervision approach to train large-scale generation models without any human annotation. This approach centers on two technical components. First, we propose a multi-level pre-training framework that incorporates massive unlabeled corpus with different quality-vs.-quantity balance at different levels. We show that models trained within this framework outperform those trained with pure human curated corpus. Second, we propose a novel self-voting-based article attention layer to extract salient information shared by multiple articles. We show that models that incorporate this layer are robust to potential noises in news stories and outperform existing baselines with or without noises. We can further enhance our model by incorporating human labels, and we show our distant supervision approach significantly reduces the demand on labeled data.

Medical image segmentation has been actively studied to automate clinical analysis. Deep learning models generally require a large amount of data, but acquiring medical images is tedious and error-prone. Attention U-Net aims to automatically learn to focus on target structures of varying shapes and sizes; thus, the name of the paper "learning where to look for the Pancreas" by Oktay et al. U-Nets are commonly used for image segmentation tasks because of its performance and efficient use of GPU memory. It aims to achieve high precision that is reliable for clinical usage with fewer training samples because acquiring annotated medical images can be resource-intensive.

"Compared to other approaches, our non-line-of-sight imaging system provides uniquely high resolutions and imaging speeds," said research team leader Christopher A. Metzler from Stanford University and Rice University. "These attributes enable applications that wouldn't otherwise be possible, such as reading the license plate of a hidden car as it is driving or reading a badge worn by someone walking on the other side of a corner." In Optica, The Optical Society's journal for high-impact research, Metzler and colleagues from Princeton University, Southern Methodist University, and Rice University report that the new system can distinguish submillimeter details of a hidden object from 1 meter away. The system is designed to image small objects at very high resolutions but can be combined with other imaging systems that produce low-resolution room-sized reconstructions. "Non-line-of-sight imaging has important applications in medical imaging, navigation, robotics and defense," said co-author Felix Heide from Princeton University.

Machine learning and deep learning are both forms of artificial intelligence. You can also say, correctly, that deep learning is a specific kind of machine learning. Both machine learning and deep learning start with training and test data and a model and go through an optimization process to find the weights that make the model best fit the data. Both can handle numeric (regression) and non-numeric (classification) problems, although there are several application areas, such as object recognition and language translation, where deep learning models tend to produce better fits than machine learning models. Machine learning algorithms are often divided into supervised (the training data are tagged with the answers) and unsupervised (any labels that may exist are not shown to the training algorithm).

Deep learning offers the promise of bypassing the procedure of manual feature engineering by learning representations in conjunction with statistical models in an end-to-end fashion. In any case, neural network architectures themselves are ordinarily designed by specialists in a painstaking, ad hoc fashion. Neural architecture search (NAS) has been touted as the way ahead for lightening this agony via automatically identifying architectures that are better than hand-planned ones. Machine learning has given some huge achievements in diverse fields as of late. Areas like financial services, healthcare, retail, transportation, and more have been utilizing machine learning frameworks somehow, and the outcomes have been promising.

OUTLINE: 0:00 - Introduction 0:43 - Talk overview 1:18 - Compute for deep learning 5:48 - Power consumption for deep learning, robotics, and AI 9:23 - Deep learning in the context of resource use 12:29 - Deep learning basics 20:28 - Hardware acceleration for deep learning 57:54 - Looking beyond the DNN accelerator for acceleration 1:03:45 - Beyond deep neural networks CONNECT: - If you enjoyed this video, please subscribe to this channel.