Please enable JavaScript.
Coggle requires JavaScript to display documents.
How Machine Discovery Can Accelerate Solutions to Society’s Big Problems -…
How Machine Discovery Can Accelerate Solutions to Society’s Big Problems
DISCOVERY OF NEW MATERIALS
Company Kebotix uses
Lab automation
Robotics
Machine learning
Computational modeling
Machines can rapidly and efficiently process enormous amounts of complex molecular or chemical data
Machines can cut research cycle from years to months
CLEAN ENERGY
Identifying stronger catalysts for the production of hydrogen from water
1,000 times faster than traditional manual laboratory methods
Much lower environmental impact than fossil fuels
Creating defect-free films for solar panels
Analysis and testing of various candidates to determine the best material
In 5 days did what usually takes 9 months
URBAN INNOVATION
Difficulties
Controlling for all variables (such as environmental conditions)
Solution
Virtual replica of an object, being, or system
Can be continuously updated with data from its physical counterpart
Serves as a sandbox environment for largescale simulations
Invaluable assets for experimentation and decision-making
Solutions
Find the technology backbone
Pooling computing resources across different innovation centers
Using public-cloud or hybrid-cloud systems.
Recalibrate commercial models
To adjust the pricing
Make ideation a human-and-machine team sport
Collaboration between humans and machines
PHARMACEUTICALS SECTOR
Company Exscientia
Made and tested only one-fifth of the usual number of candidats
Drug reached the clinical trials stage after just 12 months (it usually takes 4 and a half years)
Database search of an existing drug to treat new deseases (such as COVID-19)
Difficulties
Enormous computing power required
Solution
Runinng parallel searches across the chemical space — with staggering implications
Use elements of parallel computing and cloud system
Difficulties
Scientists may feel threatened
Up-front investment
Syncing machine discovery with commercial models