Machine Learning (ML) techniques have been rapidly adopted by smart Cyber-Physical Systems (CPS) and Internet-of-Things (IoT) due to their powerful decision-making capabilities. However, they are vulnerable to various security and reliability threats, at both hardware and software levels, that compromise their accuracy. These threats get aggravated in emerging edge ML devices that have stringent constraints in terms of resources (e.g., compute, memory, power/energy), and that therefore cannot employ costly security and reliability measures. Security, reliability, and vulnerability mitigation techniques span from network security measures to hardware protection, with an increased interest towards formal verification of trained ML models. This paper summarizes the prominent vulnerabilities of modern ML systems, highlights successful defenses and mitigation techniques against these vulnerabilities, both at the cloud (i.e., during the ML training phase) and edge (i.e., during the ML inference stage), discusses the implications of a resource-constrained design on the reliability and security of the system, identifies verification methodologies to ensure correct system behavior, and describes open research challenges for building secure and reliable ML systems at both the edge and the cloud.

Along with the great success of deep neural networks, there is also growing concern about their black-box nature. The interpretability issue affects people's trust on deep learning systems. It is also related to many ethical problems, e.g., algorithmic discrimination. Moreover, interpretability is a desired property for deep networks to become powerful tools in other research fields, e.g., drug discovery and genomics. In this survey, we conduct a comprehensive review of the neural network interpretability research. We first clarify the definition of interpretability as it has been used in many different contexts. Then we elaborate on the importance of interpretability and propose a novel taxonomy organized along three dimensions: type of engagement (passive vs. active interpretation approaches), the type of explanation, and the focus (from local to global interpretability). This taxonomy provides a meaningful 3D view of distribution of papers from the relevant literature as two of the dimensions are not simply categorical but allow ordinal subcategories. Finally, we summarize the existing interpretability evaluation methods and suggest possible research directions inspired by our new taxonomy.

As 2020 draws to a close we look back on some of the notable research developments, awards, conferences and policy in the world of artificial intelligence. In February it was reported that MIT researchers had used a machine-learning algorithm to identify a powerful new antibiotic compound. In laboratory tests, the drug, called halicin, killed many disease-causing bacteria, including some strains that had been resistant to all existing antibiotics. Progress in the field of AI in healthcare has continued apace during 2020. A lot of this work is about providing clinicians with extra tools in their armoury.

Fox News Flash top entertainment and celebrity headlines are here. Check out what's clicking today in entertainment. Lin Qi, an executive producer on "Game of Thrones" creators David Benioff and D.B. Weiss' upcoming Netflix series, died at age 39 after being poisoned. Lin was the chairman and CEO of Yoozoo Group, which he founded in 2009. The company was working with the TV creators on an adaptation of a science fiction series based on "The Three-Body Problem" trilogy of novels by Chinese author Liu Cixin.

The best book ever written about IT work or the computer industry will be 40 years old in August. Tracy Kidder's The Soul of a New Machine describes the work of Data General engineers to prototype a minicomputer, codenamed "Eagle," intended to halt the advance of the Digital Equipment Corporation's hugely successful VAX range. It won both the Pulitzer Prize and National Book Award for non-fiction, perhaps the two highest honors available for book-length journalism. Year after year, the book continues to sell and win new fans. Developers born since it was published often credit it with shaping their career choices or helping them appreciate the universal aspects of their own experiences. Soul's appeal has endured, even though what started out as a dispatch from a fast-growing firm building a piece of the future now reads as a time capsule from a lost world. Back in 1991 I read the book for an undergraduate class, typing my paper on a PC that was already more capable than Eagle yet cost 100 times less. So why are so many people still excited to relive the creation of a pitifully obsolete computer, designed by a team of obscure engineers for a long-forgotten company that never mattered very much anyway? Having spent almost 30 years now trying to take the book apart and figure out how it works, I think I have some answers. Paradoxically, the obscurity of Data General helps to explain the book's enduring power.

In some stores, sophisticated systems are tracking customers in almost every imaginable way, from recognizing their faces to gauging their age, their mood, and virtually gussying them up with makeup. The systems rarely ask for people's permission, and for the most part they don't have to. In our season 1 finale, we look at the explosion of AI and face recognition technologies in retail spaces, and what it means for the future of shopping. This episode was reported and produced by Jennifer Strong, Anthony Green, Tate Ryan-Mosley, Emma Cillekens and Karen Hao. Strong: Retailers have been using face recognition and AI tracking technologies for years. And what if you could know about the presence of violent criminals before they act? With Face First you can stop crime before it starts.] It detects faces, voices, objects and claims it can analyze behavior. But face recognition systems have a well-documented history of misidentifying women and people of color. And they're trying to sell it and impose it on the entirety of the country?] Strong: This is Representative Alexandria Ocasio-Cortez at a 2019 congressional hearing on facial recognition.

We propose a Distributional Approach to address Controlled Text Generation from pre-trained Language Models (LMs). This view permits to define, in a single formal framework, "pointwise" and "distributional" constraints over the target LM -- to our knowledge, this is the first approach with such generality -- while minimizing KL divergence with the initial LM distribution. The optimal target distribution is then uniquely determined as an explicit EBM (Energy-Based Model) representation. From that optimal representation we then train the target controlled autoregressive LM through an adaptive distributional variant of Policy Gradient. We conduct a first set of experiments over pointwise constraints showing the advantages of our approach over a set of baselines, in terms of obtaining a controlled LM balancing constraint satisfaction with divergence from the initial LM (GPT-2). We then perform experiments over distributional constraints, a unique feature of our approach, demonstrating its potential as a remedy to the problem of Bias in Language Models. Through an ablation study we show the effectiveness of our adaptive technique for obtaining faster convergence.

When answering questions about an image, it not only needs knowing what -- understanding the fine-grained contents (e.g., objects, relationships) in the image, but also telling why -- reasoning over grounding visual cues to derive the answer for a question. Over the last few years, we have seen significant progress on visual question answering. Though impressive as the accuracy grows, it still lags behind to get knowing whether these models are undertaking grounding visual reasoning or just leveraging spurious correlations in the training data. Recently, a number of works have attempted to answer this question from perspectives such as grounding and robustness. However, most of them are either focusing on the language side or coarsely studying the pixel-level attention maps. In this paper, by leveraging the step-wise object grounding annotations provided in the GQA dataset, we first present a systematical object-centric diagnosis of visual reasoning on grounding and robustness, particularly on the vision side. According to the extensive comparisons across different models, we find that even models with high accuracy are not good at grounding objects precisely, nor robust to visual content perturbations. In contrast, symbolic and modular models have a relatively better grounding and robustness, though at the cost of accuracy. To reconcile these different aspects, we further develop a diagnostic model, namely Graph Reasoning Machine. Our model replaces purely symbolic visual representation with probabilistic scene graph and then applies teacher-forcing training for the visual reasoning module. The designed model improves the performance on all three metrics over the vanilla neural-symbolic model while inheriting the transparency. Further ablation studies suggest that this improvement is mainly due to more accurate image understanding and proper intermediate reasoning supervisions.

These Visionary companies have a big idea and are well on their way to achieving it, although it isn't always an easy road for any really innovative technology. In the case of Cruise, that meant testing self driving vehicles on the streets of San Francisco, one of the hardest driving environments in the world. Some of our Visionary Awards go to companies who are opening up new market applications for robotics, such as Built Robotics in construction, Dishcraft in food services, Embark in self-driving trucks, Iron Ox in urban agriculture and Zipline in drone delivery. Some are building tools or platforms that the entire robotics industry can benefit from, such as Agility Robotics, Covariant, Formant, RobustAI and Zoox. The companies in our Good Robot Awards also show that'technologies built for us, have to be built by us'.

Self-driving vehicles would not be possible without sensors and so it's not surprising to see two small new sensors in the 2020 Silicon Valley Robotics'Good Robot' Innovation Awards, the Velabit from Velodyne and the nanoScan3 from SICK. Our other Innovation Awards go to companies with groundbreakingly new robots; from the tensegrity structure of Squishy Robotics, which will help in both space exploration and disaster response on earth, to the Dusty Robotics full scale FieldPrinter for the construction industry, and Titan from FarmWise for agriculture, which was also named one of Time's Best Inventions for 2020. Finally, we're delighted to see innovation in robotics that is affordable and collaborative enough for home robot applications, with Stretch from Hello Robot and Eve from Halodi Robotics. The Velabit, a game-changing lidar sensor, leverages Velodyne's innovative lidar technology and manufacturing partnerships for cost optimization and high-volume production, to make high-quality 3D lidar sensors readily accessible to everyone. The Velabit is smaller than a deck of playing cards, and it shatters the price barrier, costing $100.00 per sensor.