Deep Learning




Workshop on deep learning for multiphase chemistry

September 27-28, 2018

Beckman Center of the National Academy of Sciences at the Univ. of California, Irvine.

Accurate prediction of the future given a set of chemical reactants is a desirable capability in many fields. Multiphase environmental organic chemistry deals with the physicochemical transformations between gaseous, liquid and solid matter on scales ranging from nanoseconds to millennia, subatomic particles to solar systems. Artificial intelligence and machine learning lend themselves to chemical discovery in this area, in part, because traditional approaches are difficult when systems are highly non-linear and the probabilistic nature is non-trivial. Evolution of organic compounds in multiphase systems is complex as there are simultaneous sequential and parallel chemical corridors in addition to phase transitions. While the field has some understanding of the multiphase organic phenomena in relatively simple systems, little is known under realistic complex conditions. The controlling electrochemical details are relatively unknown and there is limited understanding of fundamental controlling factors of radical-driven chemistry in thin films when complex, often highly concentrated and non-ideal mixtures of water, organics and inorganics are present.

Questions to be addressed:

1.) What are the best strategies to gather and organize multiphase chemical mechanism data in sufficient quantities suitable for AI applications?

2.) What are multiphase chemistry questions currently unanswerable for which application of AI may facilitate breakthrough?

Organizing Committee: Annmarie Carlton, Pierre Baldi

Attendee Travel: The venue is conveniently located (e.g., 15 minutes from the Santa Ana airport). A suite of rooms is booked at the Ayres Hotel in Costa Mesa/Newport Beach, best reached by Uber/Lyft from SNA. We have arranged for a shuttle service to-and-from the hotel to the Beckman center. More details to follow.


September 27, 2018

(shuttle service from Ayers Hotel to Beckman Center, of the National Academy of Sciences 7:00 am) *note: The Ayers serves complimentary breakfast.

coffee, pastries at Beckman Center 7:50-8:15
logistics (Carlton) 8:15-8:25
welcome (Krnavek) 8:25-8:30
Science talks  each presentation ends with 2 slides addressing science questions 1 and 2  
Carlton (introduction and workshop goals) 8:30-8:50
Pierre Baldi 8:50-9:25
David Van Vranken 9:25 – 10:05
Connor Coley 10:05 – 10:45
break 10:45 – 11:00
Qian Yang 11:00 – 11:40
Andrey Rzhetsky 11:40 – 12:20
lunch 12:20 – 1:20
Kieron Burke 1:20 – 2:00
Evan Reed 2:00 – 2:40
Berend Smit 2:40 – 3:20
break 3:20 – 3:30
Paul Tratnyek 3:30 – 4:10
Satoshi Takahama 4:10 – 4:40
Joel Thornton 4:40 – 5:10
working dinner ‘instructions’ 5:10

**5:10 adjourn (shuttle to dinner)

September 28, 2018

(shuttle service from Ayers Hotel to Beckman Center, of the National Academy of Sciences 7:00 am) *note: The Ayers serves complimentary breakfast.

Coffee and pastries at Beckman Center 7:50-8:00
logistics (Carlton) 8:00-8:10
Group Discussions  

Are we asking the right questions?

If cost was not an issue, what would you build, study, etc.?

Have our discussions identified a topic sufficiently fundamental that any of us would interested to pursue it?

If NSF was going to have a Center call specific to the topics discussed here, what would your dream RFA say?

(groups formally present dinner and meeting discussions)

Group 1 9:30-9:45
Group 2 9:45-10:00
Group 3 10:00-10:15
break 10-:15 – 10:30
Discussion (whole group)

Synthesis of group discussions and identify common, repeated themes.

Is there expertise not present  at the workshop that should have been?

10:30 – 12:00


Lunch and adjourn 12:00-1:00