Combinatorial Optimisation Problems – from Drug Discovery to Space
summary
Fujitsu applies a Digital Annealer to map combinatorial problems into QUBO-style searches, searches libraries of up to 10 billion molecules down to ~5,000 candidates for early-stage drug discovery, then hands those to classical AI and electrostatic scoring; a dengue project produced ~3,000 candidates and moved ~15 compounds to synthesis within months. For space debris, orbital-cost neural nets feed the annealer to optimize missions, yielding thousands of times faster computations and a test run that uses 18 less fuel while finding better mission plans.
AI-written summary · about
chapters (4)
related talks
16mBeyond Patterns: The Case for Physics-Grounded AI in Drug Discovery
39mPractical Quantum Computing Today
35mCan AI solve the biggest problem in drug discovery? Time!
30mAI in Drug Discovery
21mCombinatorial Optimization of Train Timetable Fulfilment After Incidents
22mAI-powered digital transformation in healthcare and life sciences
11mEmpowering scientific innovation with HPC AI and Quantum –with Aseem Datar
17mQuantum Computing for Photon-Drug Interactions in Cancer Treatment