Mediaboss Marketing

26 November 2021

Sheppard Mullin Richter & Hampton

To print this article, all you need is to be registered or login on Mondaq.com.

The Food and Drug Administration recently sought comments on therole of transparency for artificial intelligence and machinelearning-enabled medical devices. The FDA invited comments infollow up to a recent workshop on the topic.

The workshop was part of a series of efforts the FDA has had inthis space. These include its Digital Health Center ofExcellence and a five-part Action Plan for AI andmachine-learning enabled medical devices. As part of the actionplan, the FDA indicated it wants to issue guidance on softwarelearning over time and help the industry be"patient-centered." In other words, that companies betransparent when using AI and machine learning-enabled softwarewith patients. These initiatives are especially important given theincrease in AI/ML in healthcare.

Workshop participants explored how to provide transparency. One idea proposed was using a "nutritionfact label" approach to give individuals enough information tomake informed decisions. The graphic would be similar to a foodlabel, disclosing quickly and visually the key things patientsmight want to know. (This is similar to an approach launched byApple late last year, which we discussed here.) Other agencies havelooked at machine learning and AI with similar transparencyrecommendations. We have written about those in the past, includingfor the financial servicesindustry. Advice about use of these tools has also been issuedby the FTC and the EU.

Putting it Into Practice: While the FDA continues toexplore this area, companies are reminded that the FDA (like otherregulators) expects transparency with consumers. From a privacyperspective, the workshop reminds digital health companies thisincludes telling users when AI or ML-enabled software is beingused.

The content of this article is intended to provide a generalguide to the subject matter. Specialist advice should be soughtabout your specific circumstances.

POPULAR ARTICLES ON: Technology from United States

Willkie Farr & Gallagher LLP

Last week, the Federal Trade Commission ("FTC") amended its Gramm-Leach-Bliley Act ("GLBA") Safeguards Rule (or "Rule") to, among other things, modify the scope of companies...

Bull Blockchain Law LLP

Virtual Currencies Comparative Guide for the jurisdiction of United States, check out our comparative guides section to compare across multiple countries

Read more here:
FDA Joins Other Regulators In Focus On AI And Machine Learning - Technology - United States - Mondaq News Alerts

IBM Research Europe and Thieme Chemistry have announced the first results of their collaboration which were evaluated by seven eminent synthetic chemistry experts and their research groups from China, Germany, Switzerland, New Zealand, and the USA.

Professor Dame Margaret Brimble from the University of Auckland, New Zealand comments: This innovative IBM/Thieme Chemistry platform provides an efficient tool for synthetic chemistry researchers to provide validation for their own retrosynthetic plans whilst also being presented with alternative solutions. It enables a rigorous assessment for the retrosynthetic design phase of a given synthesis which no doubt will pay dividends when the selected synthetic plan is implemented.

The partnership between IBM Research Europe and Thieme Chemistry builds on the synergies between high-quality data and state-of-the-art machine learning models for organic chemistry synthesis predictions. RXN For Chemistry, a cloud platform using artificial intelligence (AI) has recently been trained with high quality, human-curated datasets from Thiemes Science of Synthesis and Synfacts.

Organic compounds can react with each other in hundreds of thousands different ways. Experiential knowledge is key for organic chemists to avoid spending hours and hours in the laboratory with countless trials and errors. To improve synthesis planning, IBM Research and Thieme Chemistry have combined the expert human-curated datasets from Thiemes full-text resource for methods in synthetic organic chemistry, Science of Synthesis, and the reviewed content from the journal Synfacts with the artificial intelligence model called Molecular Transformer in RXN for Chemistry by IBM.

The Molecular Transformer, a neural machine translation model, was created to reliably predict the outcome of chemical reactions and was later enhanced to include retrosynthetic analysis i.e. to first determine the chemicals needed to create a given target molecule. The model has proven to be very successful at learning the information of chemical reactivity present in datasets of chemical reactions. It is, however, limited to the content and correctness of these datasets.

Science of Synthesis and Synfacts cover a wide area of reaction space. Typically, models trained on commercially available patent datasets perform poorly on many such reactions. Science of Synthesis and Synfacts have a higher quality of chemical records, reflected by a larger percentage of usable records. This consistency in Thiemes dataset facilitates the learning process of the AI models, resulting in more consistent predictions: Results show that Thieme-trained models on the RXN for Chemistry platform increase prediction accuracy by a factor of three for forward predictions, and a factor of nine for retrosynthesis.

The collaborative work between Thieme and IBM Research Europe shows the impact high-quality chemical reaction data can have on future AI chemical synthesis tools. Integrating high-quality, curated data from Science of Synthesis and Synfacts provides a unique opportunity to boost the performance of RXN for chemistry to unprecedented levels as it unleashes the entire knowledge contained in hundreds of thousands of chemical reaction records.

Professor Richmond Sarpong from the University of California, Berkeley, USA states: A sustainable future for synthesis will include minimising the number of unproductive strategies that are pursued by running only those reactions that lead to a productive end. This is only possible through the marrying of computer designed and human-designed efforts, which makes this collaboration with IBM and Thieme Chemistry exciting.

Also involved in testing the retrained models were Professor Alois Frstner (MPI Mlheim, Germany), Professor Karl Gademann and Professor Cristina Nevado (University of Zurich, Switzerland), Professor Ang Li (Shanghai Institute of Organic Chemistry, China), Professor Dirk Trauner (New York University, USA) and their research groups.

View post:
IBM Research and Thieme Chemistry partnership brings together machine learning and human-curated data - Scientific Computing World

IBL News | New York

Artificial Intelligence (AI) and machine learning, cloud computing, and 5G will be the most important technologies in 2022, according to a survey to global technology leaders from the U.S., U.K., China, India, and Brasil, conducted by IEEE.

These 350 chief technology and information officers and IT directors agreed that the pandemic accelerated the adoption of those tools.

The survey, titled The Impact of Technology in 2022 and Beyond: an IEEE Global Study, stated that 95% of tech leaders said that AI will drive the majority of innovation across nearly every industry sector in the next 1 to 5 years.

These surveyed executives consider eight areas as most benefited from 5G:

As for industry sectors impacted by technology in 2022, technology leaders cited manufacturing (25%), financial services (19%), healthcare (16%), and energy (13%).

In terms of workplace strategies and technologies, respondents say that their companies are working closely with Human Resources to implement tools for office check-in, space usage data and analytics, COVID and health protocols, employee productivity, engagement, and mental health.

Looking ahead, 81% agree that in the next five years, one-quarter of what they do will be enhanced by robots, and 77% agree that in the same time frame, robots will be deployed across their organization to enhance nearly every business function from sales and human resources to marketing and IT.

A majority of respondents agree (78%) that in the next ten years, half, or more, of what they do will be enhanced by robots. As for the deployments of robots that will most benefit humanity, according to the survey, those are manufacturing and assembly (33%), hospital and patient care (26%), and earth and space exploration (13%).

Regarding blockchain technology, the most important uses in the next year will be:

See the rest here:
AI and Machine Learning, Cloud Computing, and 5G Will Dominate in 2022 - IBL News

This artist's illustration shows the planetary system K2-138, which was discovered by citizen scientists in 2017 using data from NASA's Kepler space telescope.

After the first exoplanet was identified almost three decades earlier, in 1992, humanity has come a long way in terms of exoplanet discovery. As of today, we have spotted over 4000 validated exoplanets that revolve around their respective suns.

Exoplanets are celestial bodies that exist outside our vast solar system. Equipped with cutting-edge technology, many research groups have been identifying these exoplanets left, right and centre.

However, for the first time ever, 301 validated planets were added to the ever-growing exoplanet tally all at once!

Wondering how? The US space agency NASA reported that a new deep neural network called 'ExoMiner' was responsible for this incredible scientific feat.

The ExoMiner leverages NASA's Pleiades supercomputer and, like any deep neural network, can automatically learn a task when provided with enough data. ExoMiner is designed with various tests, properties human experts use to confirm new exoplanets, past confirmed exoplanets, and false-positive cases in mind. Thus, it could tell apart actual exoplanets from imposters, making this technology and its predictions highly reliable.

"Unlike other exoplanet-detecting machine learning programs, ExoMiner isn't a black boxthere is no mystery as to why it decides something is a planet or not," said Jon Jenkins, an exoplanet scientist at NASA's Ames Research Center in California's Silicon Valley. "We can easily explain which features in the data lead ExoMiner to reject or confirm a planet."

It is a highly tedious process to comb vast datasets from missions like Kepler, which has hundreds of stars in its range of view, each with the potential to house numerous possible exoplanets. In such cases, the ExoMiner is the perfect substitute as it reduces the burden of astronomers in sifting through data and determining what is and isn't a planet.

"When ExoMiner says something is a planet, you can be sure it's a planet," said Hamed Valizadegan, ExoMiner project lead and machine learning manager with the Universities Space Research Association at Ames. "ExoMiner is highly accurate and in some ways more reliable than both existing machine classifiers and the human experts it's meant to emulate because of the biases that come with human labelling."

NASA said that all 301 machine-validated planets were originally detected by the Kepler Science Operations Center and were promoted to planet candidate status by the Kepler Science Office. But until ExoMiner, no one was able to validate them as planets.

And while none of the newly discovered planets is thought to be Earth-like or in their parent stars' habitable zones, they share some traits with the rest of the verified exoplanet population in our galaxy.

According to Jon Jenkins, the 301 discoveries will help researchers better understand planets and solar systems beyond our own and what makes ours so unique.

**

For weather, science, and COVID-19 updates on the go, download The Weather Channel App (on Android and iOS store). It's free!

Read this article:
ExoMiner Goes Planet Hunting! NASA's Machine Learning Network Validates 301 New Exoplanets at One Go | The Weather Channel - Articles from The Weather...

Paid Post Theres no doubt that machine learning has massive potential for improving the development, delivery, and operation of public services, whether thats delivering insights into disease proliferation, enabling predictive maintenance, or identifying fraud.

This can seem to be a complex technology. But if the principles behind machine learning can be intimidating, they are nowhere near as intimidating as the consequences of getting it wrong and generating questionable or even positively harmful outcomes.

So, whether your organisation is preparing for its first journey with machine learning, or has already implemented the technology, it pays to step back and take a broader look.

And we have something that can help you in this process, in the shape of Machine Learning Reloaded, an in-depth dive into the principles and applications of machine learning in public services.

This concise but info-packed report is part of the Perspectives series from our chums at global smart software specialists Civica, with their latest volume produced in association with the UKRI Centre for Doctoral Training in Accountable, Responsible and Transparent Artificial Intelligence (ART-AI) at the University of Bath.

Machine Learning Reloaded gives you a crash course in the principles behind the tech, helping you understand what it is, what it isnt, and what it can potentially do.

It also provides an in-depth examination of how machine learning is making a difference across the full range of public services, including local government, health and care, government and justice, housing, and education.

With best use cases, and explanations of key applications, it will take you direct to a range of other resources showing how public bodies have already put the technology to work.

As well as whetting your appetite, it provides you with a template for planning your own machine learning projects, from choosing your data, selecting your tools, and choosing the right personnel and partners while guiding you on how to do all this ethically and responsibly.

Civicas NorthStar lab has already run the ruler over Chatbots, and Immersive Technologies. You can find out more and explore the rest of the Perspectives from Civica series at http://www.civica.com/perspectives. If you want to know where cutting edge technology is taking public services, this is the place to start.

See the rest here:
Machine learning can improve your public services. Are you ready to take the red pill? - The Register