Adapting Liquid ML Networks

Adapting ML networks over time is an important idea.  How much better will this be than simply recreating the networks with new data?  Automating maintenance would make sure it is done.  And note too the claim of 'exploring degrees of complexity' ... what is the value of that towards better or more resilient solutions?   Scaling down towards richer solutions?

'Liquid' ML System Adapts to Changing Conditions

MIT News, Daniel Ackerman, January 28, 2021

A team of researchers from the Massachusetts Institute of Technology (MIT), the Institute of Science and Technology Austria, and the Vienna University of Technology in Austria has developed flexible algorithms, also known as "liquid" networks, that continually alter underlying equations to adapt to new data inputs. Unlike most neural networks, whose behaviors are fixed after training, a liquid network can adapt to the variability of real-world systems and is more resilient to unexpected or noisy data. Said MIT's Ramin Hasani, "Just changing the representation of a neuron, you can really explore some degrees of complexity you couldn't explore otherwise. The model itself is richer in terms of expressivity." The network outperformed other state-of-the-art time series algorithms by a few percentage points in predicting future values in datasets. Said Hasani, "Everyone talks about scaling up their network. We want to scale down, to have fewer but richer nodes." ... 

Adapting Kiosks for the Pandemic

 Another are we addressed in the innovation centers, kiosks.  Here a good article about how the Pandemic is making us think more about their useful operation.

How the Pandemic Is Driving Kiosk Adaption

Peter Collier, SCB Contributor  in SCM

The COVID-19 pandemic threw the world for a loop. Virtually overnight, everything about the way businesses operated changed. Now, as they come to terms with new operating environments, many are looking for ways to deliver a good employee and customer experience, while improving safety and reducing costs.

Kiosks, the blanket term for self-service devices that automate transactions, can be a key part of the solution. Every day, new kiosks are being deployed in a variety of settings, including healthcare, retail, banks, food service, government agencies, school campuses, and more.

Kiosks for ticket purchase at movie theaters, airport check ins and wayfinding in malls are all familiar, but there are many other uses — especially now. For example, CleanslateUV is putting its sanitizers in department stores and restaurants so employees and customers can sanitize phones, keys and wallets when they enter and leave. Metrolinx is installing personal protective equipment (PPE) vending machines at transit hubs to ensure that passengers have access to masks and other safety items they may need to ride public transit.

And it doesn’t stop at PPE. Pharmacies are increasingly using kiosks to store and dispense drugs safely, and there’s a pilot program in Vancouver testing opioid dispensers from Signifi, using biometrics to deliver safe doses to marginalized communities.  ... " 

Bain on Adapting for Back to Work

Mor ecomments on the process and implications of getting back to work.

Back to Work
Recovery from Covid-19 will require resilience as companies advance, retreat, adapt—and repeat as necessary.

By Hernan Saenz, Dunigan O'Keeffe, Karen Harris, Tim van Biesen, John Hazan, Dianne Ledingham and Nate Anderson   in Bain

Recoveries have always mattered in business. Some of the biggest shifts in market share occur coming out of downturns, when new industry leaders—and new industries—often emerge. Rarely, though, has recovery meant putting people’s lives at risk.   

Yet that’s where we are today.

The post-Covid-19 world will accelerate some existing trends and create new ones, and all business models will have to evolve in order to grow and thrive. But there’s no way to accurately predict the coming year, and it’s a dangerous mistake to rely too heavily on forecasts, which have to be complemented with highly adaptive and resilient business models. ... "