Loading...

ÇмúÇà»ç

Korean AI Association

  >   ÇмúÇà»ç   >   ±¹³»Çмú´ëȸ

±¹³»Çмú´ëȸ

¿¬»ç ¹× ÃÊ·Ï (20ÀÏ)
 
À±¼¼¿µ ±³¼ö (Çѱ¹°úÇбâ¼ú¿ø)
 
Title: ML Basics/Optimization for ML
 

Abs

º» °­ÀÇ¿¡¼­´Â ±âº»ÀûÀÎ ±â°èÇнÀÀÇ °³³ä¿¡ ´ëÇÏ¿© ¼³¸íÇϸç ÇнÀÀ» À§ÇÑ ÃÖÀûÈ­ ¹æ¹ýÀ» ´Ù·é´Ù. ±â°èÇнÀÀ̶õ µ¥ÀÌÅ͸¦ ¹ÙÅÁÀ¸·Î ÁøÇàµÇ´Â ÇнÀÀ̸ç ÇнÀÀ» À§ÇÏ¿© ¸ñÀûÇÔ¼ö¸¦ ¼³Á¤ÇÏ°í ÃÖÀûÈ­ °úÁ¤À» °ÅÃÄ¾ß ÇÑ´Ù. ÀÌ·¯ÇÑ ÃÖÀûÈ­ °úÁ¤¿¡¼­ ¿ì¸®°¡ °í·ÁÇØ¾ß ÇÏ´Â °Í°ú °æ»ç ÇÏ°­¹ýÀ» ½ÃÀÛÀ¸·Î ±âº»ÀûÀÎ ÃÖÀûÈ­ ¹æ¹ý·ÐÀ» º» °­ÀǸ¦ ÅëÇÏ¿© ¹è¿ì°Ô µÈ´Ù. ƯÈ÷ ÃÖ±Ù ±â°èÇнÀ ¹ßÀüÀÇ °¡Àå Áß¿äÇÑ µö·¯´× ÇнÀ¿¡¼­ È¿°úÀûÀÎ ÃÖÀûÈ­ ¹æ¹ýÀÌ ¹«¾ùÀÎÁö¸¦ Á¶±Ý ´õ ÀÚ¼¼ÇÏ°Ô ¼Ò°³ÇÒ ¿¹Á¤ÀÌ´Ù. ¸¶Áö¸·À¸·Î °æ»çÇÏ°­¹ýÀÌ ºÒ°¡´ÉÇÑ »óȲ¿¡¼­ ¿ì¸®°¡ ½ÃµµÇÒ ¹æ¹ýµéÀ» ¼³¸íÇÏ¸ç °­ÀǸ¦ ¸¶¹«¸®ÇÒ ¿¹Á¤ÀÌ´Ù. 

 

Bio

- KAIST ±èÀçöAI´ëÇпø ±³¼ö (2019~)
- KAIST »ê¾÷¹×½Ã½ºÅÛ°øÇаú ±³¼ö (2017~)
- ¹Ì±¹ ·Î½º¾Ë¶ó¸ð½º ±¹°¡¿¬±¸¼Ò ¹Ú»çÈÄ¿¬±¸¿ø (2016~2017)
- ¿µ±¹ Microsoft Research ¹æ¹®¿¬±¸¿ø (2015~2016)
- ÇÁ¶û½º MSR-INRIA joint research center ¹Ú»çÈÄ¿¬±¸¿ø (2014~2015)
- ½º¿þµ§ KTH ¹Ú»çÈÄ¿¬±¸¿ø (2013~2014)
- KAIST Àü±â¹×ÀüÀÚ°øÇÐ ¹Ú»ç (2012)
- KAIST Àü±â¹×ÀüÀÚ°øÇÐ Çлç (2006)  

 

 

 


 
ÀÌ»óÇÐ ±³¼ö  (¼­¿ï´ëÇб³)
 
Title: Causal Inference
 

Abs

An intelligent agent must equip with an ability to utilize available data to make rational decisions that will bring about the desired result. To do this effectively, the agent should be able to infer the effects of its actions, not just estimating associations presented in the given data. Researchers in the field of AI who focus on causality have long examined how to formally represent observations and experiments and how to mathematically derive the consequences of actions. In this tutorial, I will introduce causal inference and other closely related research problems.
 

Bio

Sanghack Lee is an Assistant Professor at the Graduate School of Data Science at Seoul National University. Prior to that, he was an Associate Research Scientist at Columbia University, working with Professor Elias Bareinboim on the intersection of causal inference and AI. Lee received his Ph.D. in Information Sciences and Technology from Pennsylvania State University. His research focuses on developing techniques for unifying data sets from various conditions to answer causal questions, and creating decision-making algorithms that incorporate causal domain knowledge. He was the recipient of a Best Paper Award at the UAI conference in 2019. 

 


 

¿À¹Îȯ ±³¼ö(¼­¿ï´ëÇб³)
 
Title: Efficient Reinforcement Learning
 

Abs

º» °­Á¿¡¼­´Â °­È­ÇнÀ ¹®Á¦ÀÇ Æ¯¼º¿¡ ´ëÇؼ­ ¹è¿ì°í, ÃÖÀûÀÇ Á¤Ã¥(policy)À» ÇнÀÇÏ´Â °úÁ¤ Áß¿¡ ¹ß»ýÇÏ´Â °­È­ÇнÀÀÇ exploration-exploitation tradeoff¸¦ ¼Ò°³Çϸç, À̸¦ ÇØ°áÇϱâ À§ÇÑ È¿À²ÀûÀÎ °­È­ÇнÀ ¹æ¹ý·Ð¿¡ ´ëÇØ ¼Ò°³ÇÑ´Ù. 

 

Bio

¿À¹Îȯ ±³¼ö´Â Columbia ´ëÇб³¿¡¼­ ¼öÇÐ/Åë°è Çлç, Operations Research ¼®»ç ¹× ¹Ú»ç ÇÐÀ§¸¦ ¹Þ¾Ò´Ù. 2020³â 9¿ùºÎÅÍ ¼­¿ï´ëÇб³ µ¥ÀÌÅÍ»çÀ̾𽺴ëÇпø¿¡ ÀçÁ÷ ÁßÀ̸ç, °­È­ÇнÀ, ¹êµ÷ ¾Ë°í¸®Áò, ¹× Åë°èÀû ±â°èÇнÀ ºÐ¾ßÀÇ ¿¬±¸¸¦ ¼öÇàÇÏ°í ÀÖ´Ù.