Aspect Sentiment Triplet Extraction (ASTE) is the task of extracting triplets of aspect terms, their associated sentiments, and the opinion terms that provide evidence for the expressed sentiments. Previous approaches to ASTE usually simultaneously extract all three components or first identify the aspect and opinion terms, then pair them up to predict their sentiment polarities. In this work, we present a novel paradigm, ASTE-RL, by regarding the aspect and opinion terms as arguments of the expressed sentiment in a hierarchical reinforcement learning (RL) framework. We first focus on sentiments expressed in a sentence, then identify the target aspect and opinion terms for that sentiment. This takes into account the mutual interactions among the triplet's components while improving exploration and sample efficiency. Furthermore, this hierarchical RLsetup enables us to deal with multiple and overlapping triplets. In our experiments, we evaluate our model on existing datasets from laptop and restaurant domains and show that it achieves state-of-the-art performance. The implementation of this work is publicly available at https://github.com/declare-lab/ASTE-RL.

We present VESPA, an intentionally simple yet novel zero-shot system for layout, locale, and domain agnostic document extraction. In spite of the availability of large corpora of documents, the lack of labeled and validated datasets makes it a challenge to discriminatively train document extraction models for enterprises. We show that this problem can be addressed by simply transferring the information extraction (IE) task to a natural language Question-Answering (QA) task without engineering task-specific architectures. We demonstrate the effectiveness of our system by evaluating on a closed corpus of real-world retail and tax invoices with multiple complex layouts, domains, and geographies. The empirical evaluation shows that our system outperforms 4 prominent commercial invoice solutions that use discriminatively trained models with architectures specifically crafted for invoice extraction. We extracted 6 fields with zero upfront human annotation or training with an Avg. F1 of 87.50.

"In God we trust, all others must bring data." The "data economy" has been underway for a couple of decades now, where data has been recognized as an asset vital to the business. Enterprises over time have developed robust processes, policies, organizations, and platforms to manage, monitor, and measure the quality, relevancy and availability of data for business needs at every step of a data life cycle. In the modern era, the "Economics of Artificial Intelligence (AI)" is what revolutionizes business today. AI is perhaps best applied when it replaces human intelligence.

Talkdesk, a provider of cloud-based contact center software, announced $230 million in new Series D funding that more than triples the company's valuation to $10 billion, Talkdesk founder CEO Tiago Paiva confirmed to TechCrunch. New investors Whale Rock Capital Management, TI Platform Management and Alpha Square Group came on board for this round and were joined by existing investors Amity Ventures, Franklin Templeton, Top Tier Capital Partners, Viking Global Investors and Willoughby Capital. Talkdesk uses artificial intelligence and machine learning to improve customer service for midmarket and enterprise businesses. It counts over 1,800 companies as customers, including IBM, Acxiom, Trivago and Fujitsu. "The global pandemic was a big part of how customers interact and how we interacted with our customers, all working from home," Paiva said.

All the sessions from Transform 2021 are available on-demand now. Talkdesk, which provides an enterprise contact center platform, today announced that it raised $230 million at a post-money valuation of $10 billion. The round, which came from Whale Rock Capital Management, TI Platform Management, Alpha Square Group, Amity Ventures, Franklin Templeton, Top Tier Capital Partners, Viking Global Investors, and Willoughby Capital, brings the company's total raised to $498 million to date. Over the past several years, businesses have increasingly turned to cloud-based contact centers to address budding customer service challenges. The pandemic accelerated that move -- service conveniences were put in place out of necessity, which gave customers more options for interacting with companies. For example, 78% of contact centers in the U.S. now intend to deploy AI in the next 3 years, according to Canam Research.

At first glance, a recently granted South African patent relating to a "food container based on fractal geometry" seems fairly mundane. The innovation in question involves interlocking food containers that are easy for robots to grasp and stack. On closer inspection, the patent is anything but mundane. That's because the inventor is not a human being – it is an artificial intelligence (AI) system called DABUS. DABUS (which stands for "device for the autonomous bootstrapping of unified sentience") is an AI system created by Stephen Thaler, a pioneer in the field of AI and programming.

Useful information (UI) is an elusive concept in neural networks. A quantitative measurement of UI is absent, despite the variations of UI can be recognized by prior knowledge. The communication bandwidth of feature maps decreases after downscaling operations, but UI flows smoothly after training due to lower Nyquist frequency. Inspired by the low-Nyqusit-frequency nature of UI, we propose the use of spectral roll-off points (SROPs) to estimate UI on variations. The computation of an SROP is extended from a 1-D signal to a 2-D image by the required rotation invariance in image classification tasks. SROP statistics across feature maps are implemented as layer-wise useful information estimates. We design sanity checks to explore SROP variations when UI variations are produced by variations in model input, model architecture and training stages. The variations of SROP is synchronizes with UI variations in various randomized and sufficiently trained model structures. Therefore, SROP variations is an accurate and convenient sign of UI variations, which promotes the explainability of data representations with respect to frequency-domain knowledge.

The recent developments in Artificial Intelligence (AI) technologies challenge educators and educational institutions to respond with curriculum and resources that prepare students of all ages with the foundational knowledge and skills for success in the AI workplace. Research on AI Literacy could lead to an effective and practical platform for developing these skills. We propose and advocate for a pathway for developing AI Literacy as a pragmatic and useful tool for AI education. Such a discipline requires moving beyond a conceptual framework to a multi-level competency model with associated competency assessments. This approach to an AI Literacy could guide future development of instructional content as we prepare a range of groups (i.e., consumers, co-workers, collaborators, and creators). We propose here a research matrix as an initial step in the development of a roadmap for AI Literacy research, which requires a systematic and coordinated effort with the support of publication outlets and research funding, to expand the areas of competency and assessments.

Deep learning is gaining instant popularity in computer aided diagnosis of COVID-19. Due to the high sensitivity of Computed Tomography (CT) to this disease, CT-based COVID-19 detection with visual models is currently at the forefront of medical imaging research. Outcomes published in this direction are frequently claiming highly accurate detection under deep transfer learning. This is leading medical technologists to believe that deep transfer learning is the mainstream solution for the problem. However, our critical analysis of the literature reveals an alarming performance disparity between different published results. Hence, we conduct a systematic thorough investigation to analyze the effectiveness of deep transfer learning for COVID-19 detection with CT images. Exploring 14 state-of-the-art visual models with over 200 model training sessions, we conclusively establish that the published literature is frequently overestimating transfer learning performance for the problem, even in the prestigious scientific sources. The roots of overestimation trace back to inappropriate data curation. We also provide case studies that consider more realistic scenarios, and establish transparent baselines for the problem. We hope that our reproducible investigation will help in curbing hype-driven claims for the critical problem of COVID-19 diagnosis, and pave the way for a more transparent performance evaluation of techniques for CT-based COVID-19 detection.

Nowadays, technological advances have influenced all human activities, creating new dynamics and ways of communication. In this context, some artists have incorporated these advances in their creative process, giving rise to unique aesthetic expressions referred to in the literature as Generative Art, which is characterized by assigning part of the creative process to a system that acts with certain autonomy (Galanter, 2003). This research work introduces a computational system for creating generative art using a Brain-Computer Interface (BCI) which portrays the user's brain activity in a digital artwork. In this way, the user takes an active role in the creative process. In aims of showing that the proposed system materializes in an artistic piece the user's mental states by means of a visual and sound representation, several tests are carried out to ensure the reliability of the BCI device sent data. The generated artwork uses brain signals and concepts of geometry, color and spatial location to give complexity to the autonomous construction. As an added value, the visual and auditory production is accompanied by an olfactory and kinesthetic component which complements the art pieces providing a multimodal communication character.