Check out this awesome Embedding visualization toolbox by TensorFlow: https://lnkd.in/dXw2FV25! I feel that one of the silent heroes in the emergence of Data Science in the past decade has been the expansion of data visualization tools. It is less glamorous than many Deep Learning discoveries, but I am not sure we would be we are without those. One heavy influence on the world of data science has been the"The Grammar of Graphics" book by Leland Wilkinson in 2005 (https://lnkd.in/dWQRGwUD) This led to the creation of Ggplot2 in 2007 (https://lnkd.in/dEmQ4juJ).

Welcome to our October 2022 monthly digest, where you can catch up with any AIhub stories you may have missed, get the low-down on recent events, and much more. This month, we learn about a Nigerian sign language dataset, hear from researchers working on different robotic control projects, and dig into the latest governmental AI policies. Steven Kolawole created a pioneering dataset for Nigerian sign language, in collaboration with a TV sign language broadcaster and two schools in Nigeria. He used this dataset of over 8000 images to create a model to convert sign language to text or speech. In this interview, Steven told us about the goals of this research, his methodology, and how the work has inspired research in other languages.

In September 2021, the "One Hundred Year Study on Artificial Intelligence" project (AI100) issued the second report of its planned long-term periodic assessment of artificial intelligence (AI) and its impact on society. It was written by a panel of 17 study authors, each of whom is deeply rooted in AI research, chaired by Michael Littman of Brown University. The report, entitled "Gathering Strength, Gathering Storms," answers a set of 14 questions probing critical areas of AI development addressing the major risks and dangers of AI, its effects on society, its public perception and the future of the field. The report concludes that AI has made a major leap from the lab to people's lives in recent years, which increases the urgency to understand its potential negative effects. The questions were developed by the AI100 Standing Committee, chaired by Peter Stone of the University of Texas at Austin, consisting of a group of AI leaders with expertise in computer science, sociology, ethics, economics, and other disciplines.

As language models grow in popularity, it becomes increasingly important to clearly measure all possible markers of demographic identity in order to avoid perpetuating existing societal harms. Many datasets for measuring bias currently exist, but they are restricted in their coverage of demographic axes and are commonly used with preset bias tests that presuppose which types of biases models can exhibit. In this work, we present a new, more inclusive bias measurement dataset, HolisticBias, which includes nearly 600 descriptor terms across 13 different demographic axes. HolisticBias was assembled in a participatory process including experts and community members with lived experience of these terms. These descriptors combine with a set of bias measurement templates to produce over 450,000 unique sentence prompts, which we use to explore, identify, and reduce novel forms of bias in several generative models. We demonstrate that HolisticBias is effective at measuring previously undetectable biases in token likelihoods from language models, as well as in an offensiveness classifier. We will invite additions and amendments to the dataset, which we hope will serve as a basis for more easy-to-use and standardized methods for evaluating bias in NLP models.

I have never lived in Los Angeles, but I have probably spent more time thinking about L.A. than any other city that I haven't resided in. This is partly the fault of Hollywood, of Ice Cube and The White Album, of Curb Your Enthusiasm and Party Down, of the despised Lakers, but it's mostly the fault of Mike Davis. Davis, the historian and urban theorist who died on Tuesday, was probably my favorite writer about cities that I have ever read. He didn't only write about L.A., not by a long shot, but L.A. was his Beatrice, his Dark Lady. Every time I visit Los Angeles Davis' work floods through my brain, often down to specific words, phrases, and sentences.

Welcome to your official daily AI research tl;dr (often with code and news) for AI professionals where I share the most exciting papers I find daily, along with a one-liner summary to help you quickly determine if the article (and code) is worth investigating. "We propose a novel method for pre-training text-toimage generation model on image-only datasets." "We present our techniques to train a) a policy that can perform robust dexterous manipulation on an anthropomorphic robot hand and b) a robust pose estimator suitable for providing reliable real-time information on the state of the object being manipulated." The best paper award goes to Xingjian Zhen, Zihang Meng, Rudrasis Chakraborty, Vikas Singh for their paper called "On the Versatile Uses of Partial Distance Correlation in Deep Learning". "In this paper, we revisit a (less widely known) from statistics, called distance correlation (and its partial variant), designed to evaluate correlation between feature spaces of different dimensions."

We are pleased to announce that API4AI has been integrated into Eden AI API. API4AI is a cloud-native computer vision & AI platform for startups, enterprises and individual developers. They build their APIs on a complete cloud technology stack which provides full operability, scalability and stable uptime. API4AI's goal is to create out-of-the-box self-contained AI solutions that can easily be integrated into any application with just a few simple steps. Eden AI offers multiple AI APIs on its platform amongst several technologies.

Recent metrics for natural language generation rely on pre-trained language models, for instance BERTScore, BLEURT, and COMET. These metrics achieve a high correlation with human evaluations on standard benchmarks. However, it is unclear how these metrics perform for styles and domains that aren't well represented in their training data. In other words, are these metrics robust? The authors found that BERTScore isn't robust to character-level perturbations.

Capturing the similarities between human language units is crucial for explaining how humans associate different objects, and therefore its computation has received extensive attention, research, and applications. With the ever-increasing amount of information around us, calculating similarity becomes increasingly complex, especially in many cases, such as legal or medical affairs, measuring similarity requires extra care and precision, as small acts within a language unit can have significant real-world effects. My research goal in this thesis is to develop regression models that account for similarities between language units in a more refined way. Computation of similarity has come a long way, but approaches to debugging the measures are often based on continually fitting human judgment values. To this end, my goal is to develop an algorithm that precisely catches loopholes in a similarity calculation. Furthermore, most methods have vague definitions of the similarities they compute and are often difficult to interpret. The proposed framework addresses both shortcomings. It constantly improves the model through catching different loopholes. In addition, every refinement of the model provides a reasonable explanation. The regression model introduced in this thesis is called progressively refined similarity computation, which combines attack testing with adversarial training. The similarity regression model of this thesis achieves state-of-the-art performance in handling edge cases.

We introduce a new open information extraction (OIE) benchmark for pre-trained language models (LM). Recent studies have demonstrated that pre-trained LMs, such as BERT and GPT, may store linguistic and relational knowledge. In particular, LMs are able to answer ``fill-in-the-blank'' questions when given a pre-defined relation category. Instead of focusing on pre-defined relations, we create an OIE benchmark aiming to fully examine the open relational information present in the pre-trained LMs. We accomplish this by turning pre-trained LMs into zero-shot OIE systems. Surprisingly, pre-trained LMs are able to obtain competitive performance on both standard OIE datasets (CaRB and Re-OIE2016) and two new large-scale factual OIE datasets (TAC KBP-OIE and Wikidata-OIE) that we establish via distant supervision. For instance, the zero-shot pre-trained LMs outperform the F1 score of the state-of-the-art supervised OIE methods on our factual OIE datasets without needing to use any training sets. Our code and datasets are available at https://github.com/cgraywang/IELM