In this paper, a new reinforcement learning (RL) method known as the method of temporal differential is introduced. Compared to the traditional temporal-difference learning method, it plays a crucial role in developing novel RL techniques for continuous environments. In particular, the continuous-time least squares policy evaluation (CT-LSPE) and the continuous-time temporal-differential (CT-TD) learning methods are developed. Both theoretical and empirical evidences are provided to demonstrate the effectiveness of the proposed temporal-differential learning methodology.

The recent advancement in Unmanned Aerial Vehicles (UAVs) in terms of manufacturing processes, and communication and networking technology has led to a rise in their usage in civilian and commercial applications. The regulations of the Federal Aviation Administration (FAA) in the US had earlier limited the usage of UAVs to military applications. However more recently, the FAA has outlined new enforcement that will also expand the usage of UAVs in civilian and commercial applications. Due to being deployed in open atmosphere, UAVs are vulnerable to being lost, destroyed or physically hijacked. With the UAV technology becoming ubiquitous, various issues in UAV networks such as intra-UAV communication, UAV security, air data security, data storage and management, etc. need to be addressed.

Learning accurate drug representation is essential for tasks such as computational drug repositioning and prediction of drug side-effects. A drug hierarchy is a valuable source that encodes human knowledge of drug relations in a tree-like structure where drugs that act on the same organs, treat the same disease, or bind to the same biological target are grouped together. However, its utility in learning drug representations has not yet been explored, and currently described drug representations cannot place novel molecules in a drug hierarchy. Here, we develop a semi-supervised drug embedding that incorporates two sources of information: (1) underlying chemical grammar that is inferred from molecular structures of drugs and drug-like molecules (unsupervised), and (2) hierarchical relations that are encoded in an expert-crafted hierarchy of approved drugs (supervised). We use the Variational Auto-Encoder (VAE) framework to encode the chemical structures of molecules and use the knowledge-based drug-drug similarity to induce the clustering of drugs in hyperbolic space. The hyperbolic space is amenable for encoding hierarchical concepts. Both quantitative and qualitative results support that the learned drug embedding can accurately reproduce the chemical structure and induce the hierarchical relations among drugs. Furthermore, our approach can infer the pharmacological properties of novel molecules by retrieving similar drugs from the embedding space. We demonstrate that the learned drug embedding can be used to find new uses for existing drugs and to discover side-effects. We show that it significantly outperforms baselines in both tasks.

Adversarial examples are firstly investigated in the area of computer vision: by adding some carefully designed ''noise'' to the original input image, the perturbed image that cannot be distinguished from the original one by human, can fool a well-trained classifier easily. In recent years, researchers also demonstrated that adversarial examples can mislead deep reinforcement learning (DRL) agents on playing video games using image inputs with similar methods. However, although DRL has been more and more popular in the area of intelligent transportation systems, there is little research investigating the impacts of adversarial attacks on them, especially for algorithms that do not take images as inputs. In this work, we investigated several fast methods to generate adversarial examples to significantly degrade the performance of a well-trained DRL- based energy management system of an extended range electric delivery vehicle. The perturbed inputs are low-dimensional state representations and close to the original inputs quantified by different kinds of norms. Our work shows that, to apply DRL agents on real-world transportation systems, adversarial examples in the form of cyber-attack should be considered carefully, especially for applications that may lead to serious safety issues.

A robustness certificate is the minimum distance of a given input to the decision boundary of the classifier (or its lower bound). For {\it any} input perturbations with a magnitude smaller than the certificate value, the classification output will provably remain unchanged. Exactly computing the robustness certificates for neural networks is difficult since it requires solving a non-convex optimization. In this paper, we provide computationally-efficient robustness certificates for neural networks with differentiable activation functions in two steps. First, we show that if the eigenvalues of the Hessian of the network are bounded, we can compute a robustness certificate in the $l_2$ norm efficiently using convex optimization. Second, we derive a computationally-efficient differentiable upper bound on the curvature of a deep network. We also use the curvature bound as a regularization term during the training of the network to boost its certified robustness. Putting these results together leads to our proposed {\bf C}urvature-based {\bf R}obustness {\bf C}ertificate (CRC) and {\bf C}urvature-based {\bf R}obust {\bf T}raining (CRT). Our numerical results show that CRT leads to significantly higher certified robust accuracy compared to interval-bound propagation (IBP) based training. We achieve certified robust accuracy 69.79\%, 57.78\% and 53.19\% while IBP-based methods achieve 44.96\%, 44.74\% and 44.66\% on 2,3 and 4 layer networks respectively on the MNIST-dataset.

Seoul – In a cramped office in eastern Seoul, Hwang Seungwon points a remote control toward a huge NASA-like overhead screen stretching across one of the walls. With each flick of the control, a colorful array of pie charts, graphs and maps reveals the search habits of thousands of South Korean senior citizens being monitored by voice-enabled "smart" speakers, an experimental remote care service the company says is increasingly needed during the coronavirus crisis. "We closely monitor for signs of danger, whether they are more frequently using search words that indicate rising states of loneliness or insecurity," said Hwang, director of a social enterprise established by SK Telecom to handle the service. Trigger words lead to a recommendation for a visit by local public health officials. As South Korea's government pushes to allow businesses to access vast amounts of personal information and to ease restrictions holding back telemedicine, tech firms could potentially find much bigger markets for their artificial intelligence and other emerging technologies.

SpaceX has sent Nasa astronauts into space in a historic mission. The Falcon 9 rocket carried the astronauts into orbit in SpaceX's Crew Dragon capsule. Minutes after launch, the rocket detached and landed safely on a drone ship while the capsule continued to carry the astronauts on to the International Space Station. The successful launch is the first time that humans have been shot into space from US soil since the Space Shuttle programme ended in 2011. And it is the first time that humans have been sent into space by a private company, a feat only previous achieved by the space agencies of the US, Russia and China.

Artificial intelligence (AI) is an increasingly important technology within the investment management industry.1 AI has been used in a variety of ways--including as the newest strategy for attempts to "beat the market" by outperforming passive index funds that are benchmarked against the S&P 500, despite the long-standing finding that index funds consistently win that contest.2 Investment advisers who use AI should consider the unique issues the technology raises in light of an adviser's fiduciary duty to its clients. In this client alert, we provide an overview of how AI is being used by investment advisers, the fiduciary duties applicable to investment advisers, and particular issues advisers should consider in designing AI-based programs, to ensure they are acting in the best interests of their clients.3 Under federal law, an investment adviser is a fiduciary to its clients.8

You may even be using one to read this article. Wi-Fi has become essential to our personal and professional lives. The smartphone and the internet we use today wouldn't have been possible without wireless communication technologies such as Wi-Fi. In 1995 if you wanted to "surf" the internet at home, you had to chain yourself to a network cable like it was an extension cord. In 1997, Wi-Fi was invented and released for consumer use.

Artificial intelligence has influenced every aspect of our daily lives. Nowadays,thousands of tech companies have developed state-of-the-art AI-powered cybersecurity defense solutions specifically designed and programmed by ethical hackers and penetration testers. The Artificial Intelligence used in such solutions helps prevent cyberattacks from even happening by predicting the potential risks. Every tech app or service we use contains at least some type of Artificial Intelligence or smart learning technology, at least in most cases. It is now of a high possibility of connecting almost every electrical device to the internet to create our own personalized smart environments, all thanks to the recently announced 5th generation speed networks and rapid advancement in machine learning. In order to improve the overall efficiency and performance of the tasks these IoT devices are set to, they have to communicate and exchange information with each other repeatedly.