Asia AI News - Good morning and good afternoon!Here's Friday's Asia AI News, a little later than usual due to work pressures.Lots of news in today's newsletter t

Researchers at the University of California at Riverside are working to teach computer vision systems what objects typically exist in close proximity to one another so that if one is altered, the system can flag it, potentially thwarting malicious interference with artificial intelligence systems. The yearlong project, supported by a nearly $1 million grant from the Defense Advanced Research Projects Agency, aims to understand how hackers target machine-vision systems with adversarial AI attacks. Led by Amit Roy-Chowdhury, an electrical and computer engineering professor at the school's Marlan and Rosemary Bourns College of Engineering, the project is part of the Machine Vision Disruption program within DARPA's AI Explorations program. Adversarial AI attacks – which attempt to fool machine learning models by supplying deceptive input -- are gaining attention. "Adversarial attacks can destabilize AI technologies, rendering them less safe, predictable, or reliable," Carnegie Mellon University Professor David Danks wrote in IEEE's Spectrum in February.

Researchers from Florida Atlantic University's College of Engineering and Computer Science have received a four-year, $1 million grant from the National Science Foundation for a project to make the master's degree in artificial intelligence (AI) accessible to high-achieving, low-income students. The accelerated five-year bachelor's degree in science and master's degree in AI program is designed to adapt curricular and co-curricular support to enable students to complete their degrees in AI, autonomous systems or machine learning, which are critically important areas needed to advance America's global competitiveness and national security. "Artificial intelligence is transforming every walk of life from business to healthcare and enabling us to rethink how we analyze data, integrate massive amounts of information and make informed decisions that impact society, the economy and governance," said Stella Batalama, Ph.D., dean of FAU's College of Engineering and Computer Science and a co-principal investigator of the grant. "This important grant from the National Science Foundation will allow us to recruit and train talented and diverse students who are economically disadvantaged and provide them with a unique opportunity to pursue graduate education in an exciting and burgeoning field." By preparing increased numbers of high-achieving, low-income students to become engineers in these fields, this project addresses the need for growing a more diverse STEM (science, technology, engineering and mathematics) research population.

Deep neural networks (DNNs) have demonstrated excellent performance on various tasks, however they are under the risk of adversarial examples that can be easily generated when the target model is accessible to an attacker (white-box setting). As plenty of machine learning models have been deployed via online services that only provide query outputs from inaccessible models (e.g., Google Cloud Vision API2), black-box adversarial attacks (inaccessible target model) are of critical security concerns in practice rather than white-box ones. However, existing query-based black-box adversarial attacks often require excessive model queries to maintain a high attack success rate. Therefore, in order to improve query efficiency, we explore the distribution of adversarial examples around benign inputs with the help of image structure information characterized by a Neural Process, and propose a Neural Process based black-box adversarial attack (NP-Attack) in this paper. Extensive experiments show that NP-Attack could greatly decrease the query counts under the black-box setting.

Hate speech is an important problem in the management of user-generated content. In order to remove offensive content or ban misbehaving users, content moderators need reliable hate speech detectors. Recently, deep neural networks based on transformer architecture, such as (multilingual) BERT model, achieve superior performance in many natural language classification tasks, including hate speech detection. So far, these methods have not been able to quantify their output in terms of reliability. We propose a Bayesian method using Monte Carlo Dropout within the attention layers of the transformer models to provide well-calibrated reliability estimates. We evaluate and visualize the introduced approach on hate speech detection problems in several languages. From the experiments performed it was observed that our approach significantly improve the hate speech detection that can not be trusted. Our approach not only improves classification performance of the state-of-the-art multilingual BERT model, but the computed reliability scores also significantly reduce the workload in the inspection of offending cases and in reannotation campaigns. The provided visualization helps to understand the borderline outcomes.

Event forecasting is a challenging, yet consequential task, as humans seek to constantly plan for the future. Existing automated forecasting approaches rely mostly on structured data, such as time-series or event-based knowledge graphs, to help predict future events. In this work, we formulate the forecasting problem as a restricted-domain, multiple-choice, question-answering (QA) task that simulates the forecasting scenario. To showcase the usefulness of this task formulation, we introduce a dataset ForecastQA, a question-answering dataset consisting of 10,392 event forecasting questions, which have been collected and verified via crowdsourcing efforts. We also present our experiments on ForecastQA using BERT-based models and find that our best model achieves 61.0\% accuracy on the dataset, which is still far behind human performance by about 18%. We hope ForecastQA will support future research efforts in bridging this gap.

Companies that incorporate Artificial Intelligence solutions in their core values reap the benefits of the technical progress and the informational revolution as we are all standing at the dawn of a new era of virtual transformation and automation. AI is indispensable in dynamic environments, requiring you to work with data on a large scale. It makes data processing faster and more efficient, as it's specifically designed to work in a data-intense environment and learn through operating large volumes of data and analyzing information. Combining Artificial Intelligence services with Cloud Computing would be something to consider in this regard as it gives instant advantages in terms of data manipulation. The cloud environment allows you to store massive volumes of data and makes it easily accessible.

WASHINGTON--Representatives Paul D. Tonko (D-NY) and Guy Reschenthaler (R-PA) have just announced the introduction of their Artificial Intelligence Education Act today, bipartisan legislation that would establish grant support within the National Science Foundation (NSF) to fund the creation of easily-accessible K-12 lesson plans for schools and educators to provide students with the tools, skills and social understanding of artificial intelligence (AI) technologies in 21st-Century Society. "The development of Artificial Intelligence has fundamentally changed the way we live and work, bringing untold potential in the fields of medical science, research and development, engineering, manufacturing and so much more," Congressman Tonko said. "By providing the resources for our children to learn about AI, we ensure that the next generation of our American workforce has the skill necessary to succeed in this rapidly growing field, thereby helping to drive innovation and economic opportunity. Our legislation will deliver the tools needed to teach AI to students across the nation. I thank Congressman Reschenthaler for his partnership as a co-lead of this bill and urge the support of my colleagues in Congress to help secure a bright future for both our students and our economy."

The government of India has recognized the potential of AI to boost the country's GDP with $957 billion by 2035. With this in mind, the Ministry of Communication and Information Technology (MeitY) is working on the national strategy on artificial intelligence. Union Minister Rao Inderjit Singh said the draft cabinet note on implementation of national strategy on AI is being steered by MeitY and the same is under examination. Singh also stated that NITI Aayog had released India's National Strategy for Artificial Intelligence (NSAI) in June 2018. It outlined the proposed efforts in research, development, adoption and skilling in AI.

With the US presidential election now 40 days away, all eyes are focused on how online conversations, in conjunction with other hallmarks of online life like viral videos, news clips, and misleading ads, will be used, and often abused, to influence people's decisions. But political discourse, of course, is just one of the ways that user-generated content on the internet is misused for toxic ends. Today, a startup that's using AI to try to tackle them all is announcing some funding. Spectrum Labs -- which has built algorithms and a set of APIs that can be used to moderate, track, flag and ultimately stop harassment, hate speech, radicalization, and some 40 other profiles of toxic behavior, in English as well as multiple other languages -- has raised $10 million in a Series A round of funding, capital that the company plans to use to continue expanding its platform. The funding is being led by Greycroft, with Wing Venture Capital, Ridge Ventures, Global Founders Capital, and Super{set} also participating.