MELLODDY: Cross-pharma Federated Learning at Unprecedented Scale Unlocks Benefits in QSAR without Compromising Proprietary Information.
Wouter HeyndrickxLewis H. MervinTobias MorawietzNoé SturmLukas FriedrichAdam ZalewskiAnastasia PentinaLina HumbeckMartijn OldenhofRitsuya NiwayamaPeter SchmidtkeNikolas FechnerJaak SimmAdam AranyNicolas DrizardRama JabalArina AfanasyevaRegis LoebShlok VermaSimon HarnqvistMatthew HolmesBalazs PejoMaria TelenczukNicholas HolwayArne DieckmannNicola RiekeFriederike ZumsandeDjork-Arné ClevertMichael KrugChristopher N. LuscombeDarren V. S. GreenPeter ErtlPeter AntalDavid MarcusNicolas Do HuuHideyoshi FujiStephen D. PickettGergely ÁcsEric BonifaceBernd BeckYax SunArnaud GohierFriedrich RippmannOla EngkvistAndreas H. GöllerYves MoreauMathieu N. GaltierAnsgar SchuffenhauerHugo CeulemansPublished in: J. Chem. Inf. Model. (2024)
Keyphrases
- information sources
- learning process
- information processing
- prior knowledge
- learning algorithm
- higher level
- reinforcement learning
- unsupervised learning
- user interaction
- learning systems
- recent advances
- background knowledge
- making decisions
- active learning
- end users
- online learning
- website
- feature selection
- neural network
- learning problems
- learned knowledge