In many artificial intelligence and computer vision systems, the same object can be observed at distinct viewpoints or by diverse sensors, which raises the challenges for recognizing objects from different, even heterogeneous views. Multi-view discriminant analysis (MvDA) is an effective multi-view subspace learning method, which finds a discriminant common subspace by jointly learning multiple view-specific linear projections for object recognition from multiple views, in a non-pairwise way. In this paper, we propose the kernel version of multi-view discriminant analysis, called kernel multi-view discriminant analysis (KMvDA). To overcome the well-known computational bottleneck of kernel methods, we also study the performance of using random Fourier features (RFF) to approximate Gaussian kernels in KMvDA, for large scale learning. Theoretical analysis on stability of this approximation is developed. We also conduct experiments on several popular multi-view datasets to illustrate the effectiveness of our proposed strategy.

Social media, especially Twitter, is being increasingly used for research with predictive analytics. In social media studies, natural language processing (NLP) techniques are used in conjunction with expert-based, manual and qualitative analyses. However, social media data are unstructured and must undergo complex manipulation for research use. The manual annotation is the most resource and time-consuming process that multiple expert raters have to reach consensus on every item, but is essential to create gold-standard datasets for training NLP-based machine learning classifiers. To reduce the burden of the manual annotation, yet maintaining its reliability, we devised a crowdsourcing pipeline combined with active learning strategies. We demonstrated its effectiveness through a case study that identifies job loss events from individual tweets. We used Amazon Mechanical Turk platform to recruit annotators from the Internet and designed a number of quality control measures to assure annotation accuracy. We evaluated 4 different active learning strategies (i.e., least confident, entropy, vote entropy, and Kullback-Leibler divergence). The active learning strategies aim at reducing the number of tweets needed to reach a desired performance of automated classification. Results show that crowdsourcing is useful to create high-quality annotations and active learning helps in reducing the number of required tweets, although there was no substantial difference among the strategies tested.

Cardiovascular Diseases (CVD) and strokes produce immense health and economic burdens globally. Coronary Artery Disease (CAD) is the most common type of cardiovascular disease. Coronary Angiography, which is an invasive treatment, is also the standard procedure for diagnosing CAD. In this work, we illustrate a Medical Decision Support System for the prediction of Coronary Artery Disease (CAD) utilizing Fuzzy Cognitive Maps (FCMs). FCMs are a promising modeling methodology, based on human knowledge, capable of dealing with ambiguity and uncertainty, and learning how to adapt to the unknown or changing environment. The newly proposed MDSS is developed using the basic notions of Fuzzy Logic and Fuzzy Cognitive Maps, with some adjustments to improve the results. The proposed model, tested on a labelled CAD dataset of 303 patients, obtains an accuracy of 78.2% outmatching several state-of-the-art classification algorithms.

This article is made possible by Intel's GameDev BOOST program -- dedicated to helping indie game developers everywhere achieve their dreams Throughout Kevin He's career in the tech and gaming industries, something has always bothered him. While rendering and other technologies evolved tremendously over time, animation woefully lagged behind. As an engineer, He wanted to find an efficient way to create more lifelike animations, and he was convinced that advanced physics simulation and AI could solve that problem. So in 2014, He struck out on his own to start DeepMotion. The goal of the San Mateo, California-based company is to provide developers with powerful software development kits (SDK) that'll allow them to create realistic animations for games and other applications.

Recent surveys, studies, forecasts and other quantitative assessments of the progress and impact of AI highlight the precarious nature of the future of work (long after the coronavirus pandemic ends), the continuing mixed attitudes of consumers about data privacy, and the possible resilience of this year's investments in AI. The IT department's need for AI talent has tripled between 2015 and 2019, but the number of AI jobs posted by IT is still less than half of that stemming from other business units; departments recruiting AI talent in high volumes include marketing, sales, customer service, finance, and research and development. By 2025, at least two of the top 10 global retailers will establish robot resource organizations to manage nonhuman workers; 77% of retailers plan to deploy AI by 2021, with the deployment of robotics for warehouse picking as the No. 1 use case [Gartner] By 2024, AI, virtual personal assistants, and chatbots will replace almost 69% of the manager's workload [Gartner] "Supervised machine learning doesn't live up to the hype. It isn't actual artificial intelligence akin to C-3PO, it's a sophisticated pattern-matching tool… Rather than seeing exponential improvements in the quality of AI performance (a la Moore's Law), we're instead seeing exponential increases in the cost to improve AI systems"--Stefan Seltz-Axmacher, founder, Starsky Robotics "…why are we holding our hands behind our back trying to build AI without mechanisms that infants have?"--Gary "We haven't really gone to great depth with deep learning yet. We've had a limited amount of training data so far. We've had limited structures with limited compute power. But the key point is that deep learning learns the concept, it learns the features. "…such capabilities [as "deepfake" transformation of the human face] were called image processing 15 years ago, but are routinely termed AI today.

5G is bringing new use cases to the forefront, one of the most prominent being machine learning empowered health care. Since respiratory infections are one of the notable modern medical concerns and coughs being a common symptom of this, a system for recognizing and diagnosing infections based on raw cough data would have a multitude of beneficial research and medical applications. In the literature, machine learning has been successfully used to detect cough events in controlled environments. In this work, we present a novel system that utilizes Convolutional Neural Networks (CNNs) to detect cough within environment audio and diagnose three potential illnesses (i.e., Bronchitis, Bronchiolitis, and Pertussis) based on their unique cough audio features. Our detection model achieves an accuracy of 90.17% and a specificity of 89.73%, whereas the diagnosis model achieves an accuracy of about 94.74% and an F1 score of 93.73%. These results clearly show that our system is successfully able to detect and separate cough events from background noise. Moreover, our single diagnosis model is capable of distinguishing between different illnesses without the need of separate models.

This paper proposes a novel framework for the segmentation of phonocardiogram (PCG) signals into heart states, exploiting the temporal evolution of the PCG as well as considering the salient information that it provides for the detection of the heart state. We propose the use of recurrent neural networks and exploit recent advancements in attention based learning to segment the PCG signal. This allows the network to identify the most salient aspects of the signal and disregard uninformative information. The proposed method attains state-of-the-art performance on multiple benchmarks including both human and animal heart recordings. Furthermore, we empirically analyse different feature combinations including envelop features, wavelet and Mel Frequency Cepstral Coefficients (MFCC), and provide quantitative measurements that explore the importance of different features in the proposed approach. We demonstrate that a recurrent neural network coupled with attention mechanisms can effectively learn from irregular and noisy PCG recordings. Our analysis of different feature combinations shows that MFCC features and their derivatives offer the best performance compared to classical wavelet and envelop features. Heart sound segmentation is a crucial pre-processing step for many diagnostic applications. The proposed method provides a cost effective alternative to labour extensive manual segmentation, and provides a more accurate segmentation than existing methods. As such, it can improve the performance of further analysis including the detection of murmurs and ejection clicks. The proposed method is also applicable for detection and segmentation of other one dimensional biomedical signals.

This paper presents a novel framework for Speech Activity Detection (SAD). Inspired by the recent success of multi-task learning approaches in the speech processing domain, we propose a novel joint learning framework for SAD. We utilise generative adversarial networks to automatically learn a loss function for joint prediction of the frame-wise speech/ non-speech classifications together with the next audio segment. In order to exploit the temporal relationships within the input signal, we propose a temporal discriminator which aims to ensure that the predicted signal is temporally consistent. We evaluate the proposed framework on multiple public benchmarks, including NIST OpenSAT' 17, AMI Meeting and HAVIC, where we demonstrate its capability to outperform state-of-the-art SAD approaches. Furthermore, our cross-database evaluations demonstrate the robustness of the proposed approach across different languages, accents, and acoustic environments.

Network embedding aims to learn low-dimensional representations of nodes while capturing structure information of networks. It has achieved great success on many tasks of network analysis such as link prediction and node classification. Most of existing network embedding algorithms focus on how to learn static homogeneous networks effectively. However, networks in the real world are more complex, e.g., networks may consist of several types of nodes and edges (called heterogeneous information) and may vary over time in terms of dynamic nodes and edges (called evolutionary patterns). Limited work has been done for network embedding of dynamic heterogeneous networks as it is challenging to learn both evolutionary and heterogeneous information simultaneously. In this paper, we propose a novel dynamic heterogeneous network embedding method, termed as DyHATR, which uses hierarchical attention to learn heterogeneous information and incorporates recurrent neural networks with temporal attention to capture evolutionary patterns. We benchmark our method on four real-world datasets for the task of link prediction. Experimental results show that DyHATR significantly outperforms several state-of-the-art baselines.

Over 20 years ago, Kearns [1998] introduced statistical queries as a framework for designing machine learning algorithms that are tolerant to noise. The statistical query model restricts a learning algorithm to ask certain types of queries to an oracle that responds with approximately correct answers. This framework has has proven useful, not only for designing noise-tolerant algorithms, but also for its connections to other noise models, for its ability to capture many of our current techniques, and for its explanatory power about the hardness of many important problems. Researchers have also found many connections between statistical queries and a variety of modern topics, including to evolvability, differential privacy, and adaptive data analysis. Statistical queries are now both an important tool and remain a foundational topic with many important questions. The aim of this survey is to illustrate these connections and bring researchers to the forefront of our understanding of this important area.