July 24-26, 2018

Boston, USA

Pre Conference Workshop A
Tuesday, 24th July, 2018

10.00am - 12.30pm
AI in Discovery to Clinic: A Workshop on Strategies & Applications
Workshop Leader: Alex Zhavoronkov, CEO, InSilico Medicine

 

  • The several DNN-based methods have been proposed for molecular de novo design and molecular feature extraction

 

  • The Generative Adversarial Networks (GANs) are being applied for molecular feature extraction using the fingerprint or string (in formats like SMILES or InChi) representation of molecular structure

 

  • The DNN-models based on GANs extended with a Reinforcement Learning (RL)-based generator could generate valid SMILES string, however the adding of objective reward function is needed to reward the generator for the whole molecular sequence

 

  •  The using of different objective reward functions in Objective-Reinforced Generative Adversarial Network for Inverse-design Chemistry (ORGANIC) makes possible to generate molecules with desired user-specified properties

 

  • The novel Adversarial Threshold Neural Computer (ATNC) architecture including a specific adversarial threshold (AT) block and using Differentiable Neural Computer (DNC) as generator outperforms the ORGANIC model in generating the valid and unique molecular structures in SMILES representation

 

 

Alex Zhavoronkov, CEO, InSilico Medicine

Dr. Zhavoronkov specializes in the development of the next-generation artificial intelligence and Blockchain technologies for drug discovery, biomarker development and aging research. At Insilico he pioneered the applications of generative adversarial networks and reinforcement learning techniques for generating the novel molecular structures with the desired properties and launched multiple research and consumer oriented biomarker systems including the popular iPANDA system and Young.AI. Prior to founding Insilico Medicine, he worked in senior roles at ATI Technologies (acquired by AMD in 2006), NeuroG, the Biogerontology Research Foundation and YLabs.AI and established AgeNet.net competitions and diversity.AI initiative. Since 2012 he published over 80 peer-reviewed research papers and books including “The Ageless Generation: How Biomedical Advances Will Transform the Global Economy”.

Pre Conference Workshop B
Tuesday, 24th July, 2018

13.30pm - 16.30pm
Artificial Intelligence, Machine Learning, and Precision Medicine: Colliding Technology & Science to Improve Personalized Patient Outcomes
Workshop Leader: Haoda Fu, Research Advisor & Enterprise Lead for Machine Learning and Artificial Intelligence Group, Eli Lilly

 

  • An overview of statistical machine learning, and artificial intelligence techniques with applications to the precision medicine, in particular to deriving optimal individualized treatment strategies for precision medicine

 

  • Covering both treatment selection and treatment transition. The treatment selection framework is based on outcome weighted classification

 

  • A deep-delve into logistic regression, support vector machine (SVM), robust SVM, and angle based classifiers for multi-category learning as well as showcasing how to modify these classification methods into outcome weighted learning algorithms for precision medicine

 

  • An introduction on reinforcement learning techniques for the purpose of  treatment transition

 

  • Algorithms, including dynamic programming for Markov Decision Process, temporal difference learning, SARSA, Q-Learning algorithms, actor-critic methods, will be covered followed by a discussion on how to use these methods for developing optimal treatment transition strategies. The techniques discussed will be demonstrated in R

 

Haoda Fu, Research Advisor & Enterprise Lead for Machine Learning and Artificial Intelligence Group, Eli Lilly

Dr. Haoda Fu is a research advisor and a stats group leader for Machine Learning, artificial Intelligence, and Digital Connected Care from Eli Lilly and Company. He is also an adjunct professor of biostatistics department, Indiana university school of medicine. Dr. Fu received his Ph.D. in statistics from University of Wisconsin – Madison in 2007 and joined Lilly after that. Since he joined Lilly, he is very active in statistics methodology research. He has more than 70 publications in the areas, such as Bayesian adaptive design, survival analysis, personalized medicine, indirect and mixed treatment comparison, joint modeling, Bayesian decision making, and drug safety evaluation for rare events. In recent years, his research area focuses on machine learning and artificial intelligence.