Mathematics is the foundation of countless sciences, allowing us to model things like planetary orbits, atomic motion, signal frequencies, protein folding, and more. Moreover, it's a valuable testbed for the ability to problem solve, because it requires problem solvers to analyze a challenge, pick out good methods, and chain them together to produce an answer. It's revealing, then, that as sophisticated as machine learning models are today, even state-of-the-art models struggle to answer the bulk of math problems correctly. A new study published by researchers at the University of California, Berkeley finds that large language models including OpenAI's GPT-3 can only complete 2.9% to 6.9% of problems from a dataset of over 12,500. The coauthors believe that new algorithmic advancements will likely be needed to give models stronger problem-solving skills.

But the big data industry has significant inertia moving into 2021. In order to give our valued readers a pulse on important new trends leading into next year, we here at insideBIGDATA heard from all our friends across the vendor ecosystem to get their insights, reflections and predictions for what may be coming. We were very encouraged to hear such exciting perspectives. Even if only half actually come true, Big Data in the next year is destined to be quite an exciting ride. The "analytic divide" is going to get worse. Like the much-publicized "digital divide" we're also seeing the emergence of an "analytic divide." Many companies were driven to invest in analytics due to the pandemic, while others have been forced to cut anything they didn't view as critical to keep the lights on – and a proper investment in analytics was, for these organizations, analytics was on the chopping block. This means that the analytic divide will further widen in 2021, and this trend will continue for ...

LA JOLLA--(March 8, 2021) Deep learning is a potential tool for scientists to glean more detail from low-resolution images in microscopy, but it's often difficult to gather enough baseline data to train computers in the process. Now, a new method developed by scientists at the Salk Institute could make the technology more accessible--by taking high-resolution images, and artificially degrading them. The new tool, which the researchers call a "crappifier," could make it significantly easier for scientists to get detailed images of cells or cellular structures that have previously been difficult to observe because they require low-light conditions, such as mitochondria, which can divide when stressed by the lasers used to illuminate them. It could also help democratize microscopy, allowing scientists to capture high-resolution images even if they don't have access to powerful microscopes. The findings were published March 8, 2021, in the journal Nature Methods.

An academic and a lawyer have teamed up to develop a robot lawyer, which, if successful, will make legal advice affordable to people from all backgrounds, while revolutionizing the legal sector. Robots could take on significant parts of a lawyer's work, reducing the costs and barriers to access to legal services for everyone, rather than just those who can afford the high costs. The project, at the University of Bradford, is initially working on a machine learning-based application to provide immigration-related legal advice, but if successful, it could be replicated across the legal sector. The project was devised by Yash Dubal, immigration lawyer and director at AY&J, and Dhaval Thakker, associate professor at the faculty of engineering and informatics at the University of Bradford. It will harness complex knowledge graph technology and deep learning algorithms to analyse case law and learn from it.

Einops, an abbreviation of Einstein-Inspired Notation for operations is an open-source python framework for writing deep learning code in a new and better way. Einops provides us with new notation & new operations. It is a flexible and powerful tool to ensure code readability and reliability with minimalist yet powerful API. In case you need convincing arguments for setting aside time to learn about einsum (https://t.co/2lA3Bsh53D) and Alex Rogozhnikov's einops (https://t.co/SY4yJAktEh). Here are a few examples to get started with Einops.

Programs like AlphaZero and GPT-3 are massive accomplishments: they represent years of sustained work solving a difficult problem. But these problems are squarely within the domain of traditional AI. Playing Chess and Go or building ever-better language models have been AI projects for decades. The common thread through these advances is applying work in one field to another area that's apparently unrelated--not sustained research at cracking a core AI problem. Using NLP to analyze mutations?

Corey Jaskolski is the founder and CEO of Synthetaic, the leading synthetic data company for impossible AI. It's a common misconception that most businesses are drowning in data and that AI is spoiled for choice when it comes to training data. The truth is that despite the big data boom, most businesses still lack the quantity of high-quality data they need, and it's holding back the development of the highest-value AI applications. Current AI research is yielding phenomenal results, but the AI beating the world's best players at Go or outperforming human lip readers are the exception, not the rule. AI keeps getting better by training on increasingly massive models, some of which have a billion tunable parameters.

GPT-3 from OpenAI has captured public attention unlike any other AI model in the 21st century. The sheer flexibility of the model in performing a series of generalized tasks with near-human efficiency and accuracy is what makes it so exciting. It has created a paradigm shift in the world of Natural Language Processing(NLP), where till now the models were trained based on the ungenralized approach to excel at one or two tasks. GPT-3 is trained by OpenAI with a generalized approach on a massive scale involving 175 billion parameters which allows it to mimic functionalities of the human brain (like GPT-3 is capable of generating text that is surprisingly human-like after only being fed a few examples of the task you want it to do). Like a human brain GPT-3 is able to learn and do things with few shots of training unlike the conventional way of training an NLP model over a large corpus, which is both difficult and time-consuming.

Odei Garcia-Garin et al. from the University of Barcelona have developed a deep learning-based algorithm able to detect and quantify floating garbage from aerial images. They also made a web-oriented application allowing users to identify these garbages, called floating marine macro-litter, or FMML, within images of the sea surface.

After rapid growth over the past few years, artificial intelligence has become one of the biggest focuses of enterprises. Well, what has made it so hot? With AI, we can design systems that learn and adapt to all the new data we collect. Just a few years ago, AI seemed to be impossible. But now, it's quickly becoming necessary and expected.