.  . .  .  . .  .  . .  .  . .  .  . .  
   .       .       .       .       .     .
     .  .    .  :8:SS @t    . .    .    
 .       .  8:S @88888@@S@;8.S    .    .  
   .  .    .t@8Xt.:t@@@88@tX%@@ .   .   . 
  .    .  .t@8@% 88   t@8@8XX%8       .   
    .     :%8@ S:XS;@  8Xt@@@8@8 . .     .
  .   . . tX :   .    :8X888@%8      .  . 
    .     S@@:     .  8%88888;8S          
  .    .  t%t;XSX8t: .%S:88 888X;  . . .  
     .   .@8S8t8XX :8tX8Xt% X88        .
  .    . XXX8%SX;X8..tt;X%:S 888t. .  .   
    .    .%@88S:   XS8X@8XS8 8S.:       . 
  .   .   SX888X88 @8888 88S88..   .    
    .   . .88888@8@8%@ 88%888 88.  .   .  
  .       .888888888tt8 8888 88@  .      .
     . .  .:888888@8888@8:8888%...   .    
  .      . :@88888.@8@X%;8 8 X     .   .  
    .  .    ;::8888 8%888S888   .         
  .       . ..;. 88X88@8888; .    . .  .  
    . .  .  .   :8@8%8@88;..   .     .    
  .          . :8S@%@@8SX@S      .      . 
     . .  .   :88X8@88:8XX8. . .   . .    
  .         .:888888 8X8X88@      .    .  

Wang-Zhou DAI

Can be pronounced as "won joe dye"

Ph.D.
Associate Professor
School of Intelligence Science and Technology
Nanjing University, Suzhou Campus

About

My research interest is in machine learning, a sub-field of artificial intelligence. Currently, I am interested in combining sub-symbolic machine learning and logic-based symbolic machine learning. (CV)

Research

Leveraging the power of logic reasoning in machine learning is my main focus right now. Popular machine learning techniques, such as Deep Neural Network and Statistical Learning, are good at mapping noisy sub-symbolic data (e.g. images) into symbols (e.g., labels, clusters, etc.); While symbolic machine learning techniques, such as Inductive Logic Programming and Statistical Relational Learning, are good at modelling complex (e.g., recursive) relationships in symbolic data. The two sub-areas in AI have been developed separately throughout the most of the history, resulting a huge gap between machine perception and reasoning. I am trying in various aspects to bridge the two islands, aiming at building ultra-strong machine learning systems that are human understandable, sample-efficient and applicaple to physical-world tasks.

Publications Codes

News

Teaching

Academic Service

Talks