Spatio-Temporal prediction plays a critical role in smart city construction. Jointly modeling multiple spatio-temporal tasks can further promote an intelligent city life by integrating their inseparable relationship. However, existing studies fail to address this joint learning problem well, which generally solve tasks individually or a fixed task combination. The challenges lie in the tangled relation between different properties, the demand for supporting flexible combinations of tasks and the complex spatio-temporal dependency. To cope with the problems above, we propose an Automated Spatio-Temporal multi-task Learning (AutoSTL) method to handle multiple spatio-temporal tasks jointly. Firstly, we propose a scalable architecture consisting of advanced spatio-temporal operations to exploit the complicated dependency. Shared modules and feature fusion mechanism are incorporated to further capture the intrinsic relationship between tasks. Furthermore, our model automatically allocates the operations and fusion weight. Extensive experiments on benchmark datasets verified that our model achieves state-of-the-art performance. As we can know, AutoSTL is the first automated spatio-temporal multi-task learning method.

ABSTRACT Speech-centric machine learning systems have revolutionized a number of leading industries ranging from transportation and healthcare to education and defense, fundamentally reshaping how people live, work, and interact with each other. However, recent studies have demonstrated that many speech-centric ML systems may need to be considered more trustworthy for broader deployment. Specifically, concerns over privacy breaches, discriminating performance, and vulnerability to adversarial attacks have all been discovered in ML research fields. In order to address the above challenges and risks, a significant number of efforts have been made to ensure these ML systems are trustworthy, especially private, safe, and fair. In this paper, we conduct the first comprehensive survey on speech-centric trustworthy ML topics related to privacy, safety, and fairness. In addition to serving as a summary report for the research community, we highlight several promising future research directions to inspire researchers who wish to explore further in this area.

The physical attack has been regarded as a kind of threat against real-world computer vision systems. Still, many existing defense methods are only useful for small perturbations attacks and can't detect physical attacks effectively. In this paper, we propose a random-patch based defense strategy to robustly detect physical attacks for Face Recognition System (FRS). Different from mainstream defense methods which focus on building complex deep neural networks (DNN) to achieve high recognition rate on attacks, we introduce a patch based defense strategy to a standard DNN aiming to obtain robust detection models. Extensive experimental results on the employed datasets show the superiority of the proposed defense method on detecting white-box attacks and adaptive attacks which attack both FRS and the defense method. Additionally, due to the simpleness yet robustness of our method, it can be easily applied to the real world face recognition system and extended to other defense methods to boost the detection performance.

The conventional approach to Bayesian decision-theoretic experiment design involves searching over possible experiments to select a design that maximizes the expected value of a specified utility function. The expectation is over the joint distribution of all unknown variables implied by the statistical model that will be used to analyze the collected data. The utility function defines the objective of the experiment where a common utility function is the information gain. This article introduces an expanded framework for this process, where we go beyond the traditional Expected Information Gain criteria and introduce the Expected General Information Gain which measures robustness to the model discrepancy and Expected Discriminatory Information as a criterion to quantify how well an experiment can detect model discrepancy. The functionality of the framework is showcased through its application to a scenario involving a linearized spring mass damper system and an F-16 model where the model discrepancy is taken into account while doing Bayesian optimal experiment design.

Chain-of-thought prompting has demonstrated remarkable performance on various natural language reasoning tasks. However, it tends to perform poorly on tasks which requires solving problems harder than the exemplars shown in the prompts. To overcome this challenge of easy-to-hard generalization, we propose a novel prompting strategy, least-to-most prompting. The key idea in this strategy is to break down a complex problem into a series of simpler subproblems and then solve them in sequence. Solving each subproblem is facilitated by the answers to previously solved subproblems. Our experimental results on tasks related to symbolic manipulation, compositional generalization, and math reasoning reveal that least-to-most prompting is capable of generalizing to more difficult problems than those seen in the prompts. A notable finding is that when the GPT-3 code-davinci-002 model is used with least-to-most prompting, it can solve the compositional generalization benchmark SCAN in any split (including length split) with an accuracy of at least 99% using just 14 exemplars, compared to only 16% accuracy with chain-of-thought prompting. This is particularly noteworthy because neural-symbolic models in the literature that specialize in solving SCAN are trained on the entire training set containing over 15,000 examples. We have included prompts for all the tasks in the Appendix.

Intelligent Virtual Machine (VM) provisioning is central to cost and resource efficient computation in cloud computing environments. As bootstrapping VMs is time-consuming, a key challenge for latency-critical tasks is to predict future workload demands to provision VMs proactively. However, existing AI-based solutions tend to not holistically consider all crucial aspects such as provisioning overheads, heterogeneous VM costs and Quality of Service (QoS) of the cloud system. To address this, we propose a novel method, called CILP, that formulates the VM provisioning problem as two sub-problems of prediction and optimization, where the provisioning plan is optimized based on predicted workload demands. CILP leverages a neural network as a surrogate model to predict future workload demands with a co-simulated digital-twin of the infrastructure to compute QoS scores. We extend the neural network to also act as an imitation learner that dynamically decides the optimal VM provisioning plan. A transformer based neural model reduces training and inference overheads while our novel two-phase decision making loop facilitates in making informed provisioning decisions. Crucially, we address limitations of prior work by including resource utilization, deployment costs and provisioning overheads to inform the provisioning decisions in our imitation learning framework. Experiments with three public benchmarks demonstrate that CILP gives up to 22% higher resource utilization, 14% higher QoS scores and 44% lower execution costs compared to the current online and offline optimization based state-of-the-art methods.

Google Cloud is tackling the cumbersome problem in healthcare of slow preauthorization of recommended procedures, medications, or devices. On April 13, the company introduced its AI-enabled Claims Acceleration Suite to reduce these administrative burdens and costs for health plans and providers. The prior authorization process causes burnout for physicians, notes Amy Waldron, director of Global Health Plans Strategy and Solutions for Google Cloud. In fact, 88% of physicians call it "very or extremely" burdensome, according to the Medical Group Management Association. The Centers for Medicare & Medicaid says prior authorizations take an average of 10 days.

No one wants to be falsely accused of saying or doing something that will destroy their reputation. Even more nightmarish is a scenario where, despite being innocent, the fabricated "evidence" against a person is so convincing that they are unable to save themselves. Yet thanks to a rapidly advancing type of artificial intelligence (AI) known as "deepfake" technology, our near-future society will be one where everyone is at great risk of having exactly that nightmare come true. Deepfakes -- or videos that have been altered to make a person's face or body appear to do something they did not in fact do -- are increasingly used to spread misinformation and smear their targets. Political, religious and business leaders are already expressing alarm by the viral spread of deepfakes that maligned prominent figures like former US President Donald Trump, Pope Francis and Twitter CEO Elon Musk.

In October, the White House released a 70-plus-page document called the "Blueprint for an A.I. Bill of Rights." The document's ambition was sweeping. It called for the right for individuals to "opt out" from automated systems in favor of human ones, the right to a clear explanation as to why a given A.I. system made the decision it did, and the right for the public to give input on how A.I. systems are developed and deployed. But if it did become law, it would transform how A.I. systems would need to be devised. And, for that reason, it raises an important set of questions: What does a public vision for A.I. actually look like?

Garante, the Italian data protection authority, apparently jumped the gun at the end of March when it imposed a temporary ban on ChatGPT, a chatbot that uses artificial intelligence (AI) to generate texts that seem as if they were created by humans, and computer games. The watchdog was less concerned by the use of AI -- the simulation of human intelligence by computer systems -- than by breaches of data protection legislation. Garante then told the Microsoft Corp-backed company behind ChatGPT, OpenAI, that it would have to be more transparent with its users about how their data were processed. It also said that the US company had to obtain permission from users if their data were to be used to further develop the software -- that is, to help it learn -- and that access to minors had to be filtered. In a press release, the Italian authority said that the ban would be lifted if OpenAI met these conditions by April 30.