Researchers at Facebook and New York University (NYU) claim to have developed three machine learning models that could help doctors predict how a COVID-19 patient’s condition might develop. The open-sourced models, all of which require no more than a sequence of X-rays, ostensibly predict patient deterioration up to four days in advance and the amount of supplemental oxygen (if any) a patient…
Oqton raises $40 million to apply AI to additive manufacturing
January 15, 2021
Oqton, a startup developing a factory operating system that integrates engineering software with manufacturing hardware, today raised $40 million. The company says the funds will be used to further develop Oqton’s platform while expanding its commercial partnerships in…
This AI smart crib gently bounces waking babies back to sleep
January 15, 2021
There is nothing in the world more valuable to new parents than sleep. To help with that, Cradlewise developed a smart crib that uses AI and a built-in monitor to help your little one stay snoozing for as long as possible. A sleekly designed bassinet in white and …
Lumiata, a company providing AI-powered predictive analytics for managing health care costs, has raised $14 million. The company says it will use the funds to scale its platform and invest in customer acquisition ahead of the opening of an office in Guadalajara, Mexico in 2021.
As many as 3.5 million adult hospital stays in 2017 — to the tune of nearly $34 billion — were considered potentially…
Google today launched Guest Mode, a new way to use Google Assistant on smart devices like Nest Audio and Nest Hub Max without having to sign into a Google account. First announced last fall, Guest Mode allows users to use Google Assistant-powered smart speakers and smart…
The AI arms race comes to enterprise content management
January 14, 2021
Enterprise content management (ECM) platforms that have historically been employed to manage files are, thanks to the rise of AI, about to evolve into central repositories for keeping track of relationships between a much wider range of types of data.
Fresh off raising an…
In the subfield of machine learning known as natural language processing (NLP), robustness testing is the exception rather than the norm. That’s particularly problematic in light of work showing that many NLP models leverage spurious connections that inhibit their performance outside of specific tests. One report found that 60% to 70% of answers given by NLP models were embedded somewhere in the…
Molecula, which is developing a cloud-based feature store for AI and machine learning workloads, today announced it has raised $17.6 million. The company says the proceeds will be put toward accelerating the launch of its managed cloud service and bolstering its sales and…
Nayya, an insurance benefits management platform, today announced it has closed $11 million in a series A round led by Felicis Ventures. Nayya says the funds will be put toward product research and development as it seeks to acquire new talent and customers.
Employees are…
Google trained a trillion-parameter AI language model
January 13, 2021
Parameters are the key to machine learning algorithms. They’re the part of the model that’s learned from historical training data. Generally speaking, in the language domain, the correlation between the number of parameters and sophistication has held up remarkably well. For example, OpenAI’s GPT-3 — one of the largest language models ever trained, at 175 billion parameters — can make…
