Introduction to Machine Learning
  • 2016 Fall Undergraduate Course in Computer Science and Engineering
  • Instructor: Prof. Byoung-Tak Zhang
  • TA: Sang-Woo Lee (Room: 138-417, Tel.: 02-880-5890), Hanock Kwak (Room: 302-314-1, Tel.: 02-880-1847)
  • Classroom: 302-107
  • Time: Tue & Thu, 11:00-12:15
  • Reference:
  • Evaluation:
    • Two open-book exams (50%)
    • Two miniprojects + two reports + one poster presentation (40%)
    • Participation and discussion (10%)
  • Announcement:
    • (09/25) ±â°èÇнÀ °³·Ð ¼ö¾÷ÀÇ ±³°ú¼­¿¡ ÇØ´çÇÏ´Â ÀÚ·á 1Àå°ú 2ÀåÀÌ °ø°³µÇ¾ú½À´Ï´Ù.
    • (10/04) ±â°èÇнÀ °³·Ð ¼ö¾÷ÀÇ ±³°ú¼­¿¡ ÇØ´çÇÏ´Â ÀÚ·á 3Àå°ú 4ÀåÀÌ °ø°³µÇ¾ú½À´Ï´Ù.
    • (10/04) Áß°£°í»ç´Â 10¿ù 20ÀÏÀÔ´Ï´Ù.
    • (10/06) Áß°£ ÇÁ·ÎÁ§Æ®¿¡ ´ëÇÑ ¼³¸í°ú ÀÚ·á°¡ °ø°³µÇ¾ú½À´Ï´Ù.
    • (10/13) ½Ç½À½Ç ÃâÀÔ ½ÅûÀÌ http://scard.snu.ac.kr (ÃâÀÔ°ü¸® -> ÃâÀÔµî·Ï½Åû) ¿¡¼­ °¡´ÉÇÕ´Ï´Ù.
      »ç¿ë¸ñÀû: ±â°èÇнÀ°³·Ð, »ç¿ëÀ§Ä¡: 302-311-1È£
    • (10/14) 10¿ù 18ÀÏ È­¿äÀÏÀº ÀÚÀ²ÇнÀÀ¸·Î ÈÞ°­ÀÔ´Ï´Ù.
    • (10/14) 10¿ù 31ÀÏ ¿ù¿äÀÏ Àú³á 7½Ã¿¡ º¸Ãæ ¼ö¾÷ÀÌ ÀÖ½À´Ï´Ù.
    • (10/25) Áß°£ ÇÁ·ÎÁ§Æ® ±âÇÑÀÌ 11¿ù 1ÀÏ·Î º¯°æµÇ¾ú½À´Ï´Ù.
    • (10/31) ±â°èÇнÀ °³·Ð ¼ö¾÷ÀÇ ±³°ú¼­¿¡ ÇØ´çÇÏ´Â ÀÚ·á 5Àå°ú 6ÀåÀÌ °ø°³µÇ¾ú½À´Ï´Ù.
    • (11/08) Áß°£°í»ç °á°ú°¡ °øÁöµÇ¾ú½À´Ï´Ù. [°á°ú]
    • Claim ±â°£ ¹× Àå¼Ò: 11¿ù 10ÀÏ ¸ñ¿äÀÏ Àú³á 7½Ã ~ Àú³á 8½Ã, 302µ¿ 314-1È£
    • (11/10) Áß°£ ÇÁ·ÎÁ§Æ® °á°ú°¡ °øÁöµÇ¾ú½À´Ï´Ù. [°á°ú]
    • (11/15) ±â°èÇнÀ °³·Ð ¼ö¾÷ÀÇ ±³°ú¼­¿¡ ÇØ´çÇÏ´Â ÀÚ·á 7ÀåÀÌ °ø°³µÇ¾ú½À´Ï´Ù.
    • (11/25) ±â¸» ÇÁ·ÎÁ§Æ® Á¶ Æí¼ºÀ» °øÁöÇÕ´Ï´Ù. [Æí¼º °á°ú]
    • (11/29) ±â°èÇнÀ °³·Ð ¼ö¾÷ÀÇ ±³°ú¼­¿¡ ÇØ´çÇÏ´Â ÀÚ·á 9ÀåÀÌ °ø°³µÇ¾ú½À´Ï´Ù.
    • (12/08) ±â¸»°í»ç´Â 12¿ù 13ÀÏ È­¿äÀÏ 11:00-12:30 ÀÔ´Ï´Ù.
    • ½ÃÇè ±ÔÄ¢Àº Áß°£°í»ç¿Í °°À¸¸ç, ¿ÀǺÏ, ÀüÀÚ±â±â ºÒÇãÀÔ´Ï´Ù.
    • ½ÃÇè ¹üÀ§´Â Áß°£ °í»ç ÀÌÈÄ ¹è¿î ³»¿ëÀ̸ç, ±³°ú¼­ 5Àå, 6Àå, 7Àå, 9Àå¿¡ ÇØ´çÇÕ´Ï´Ù.
    • (12/15) Claim ÀÏÁ¤Àº, 12¿ù 20ÀÏ È­¿äÀÏ, ¿ÀÀü 10½Ã, 302µ¿ 314-1È£ÀÔ´Ï´Ù.
    • (12/19) ±â¸»°í»ç °á°ú°¡ °øÁöµÇ¾ú½À´Ï´Ù. [°á°ú]
    • (12/26) ±â¸»°í»ç ÇÁ·ÎÁ§Æ® °á°ú°¡ °øÁöµÇ¾ú½À´Ï´Ù. [°á°ú]
  • Course Description:
    • Machine learning studies architectures and algorithms for automatically constructing models (or computer programs) from observed data. Machine learning systems design software not by manual programming but by automatic programming through feeding data, and thus can improve their performance on their own by observing more data. Deep learning is a class of machine learning models that employs many layers of neural network structures and, recently, has revolutionized the AI industry by solving many difficult problems, such as image recognition, speech recognition, natural language processing, self-driving cars, and autonomous robots. This course gives an introduction to machine learning and neural networks with an emphasis on deep learning models. We study the basic mathematical algorithms of various deep learning models and their applications in vision, speech, language, robotics, games, and digital media. Course attendants are expected to do their own deep learning AI project on a problem chosen by the participants.

    • Lecture Schedule
    • Week Topics Slides
      Week 1
      (9/1)
      • Artificial Intelligence and Machine Learning
       

      Week 2

      (9/6, 9/8)

      • Neural Networks and Delta Learning
      01   02  
      03   04  

      Week 3

      (9/13)

      • Multilayer Neural Networks and Backpropagation Learning
      05   06   07   08  

      Week 4

      (9/20, 9/22)

      • Deep Learning: What, Why, and How?
        09   10  

      Week 5

      (9/27, 9/29)

      • Models of Deep Learning
      Practice 01 (Python)  
      Practice 02 (Tensorflow)  

      Week 6

      (10/4, 10/6)

      • Convolutional Neural Networks (CNN)
      11   12  
           

      Week 7

      (10/11, 10/13)

      • Applications of CNN
      Using Neural Network  
      Practice 03 (Tensorflow)   [code]

      Week 8

      (10/18, 10/20)

      • Mid-term Exam
       

      Week 9

      (10/25, 10/27)

      • Deep Belief Networks (DBN)
         

      Week 10

      (11/1, 11/3)

      • Deep Hypernetworks
      • Applications of DBN
      (10/31)  
      (11/1) DHN for KidsVideo  
      (11/3) Practice 04 (Tensorflow,DBN)

      Week 11

      (11/8, 11/10)

      • Recurrent Neural Networks (RNN)
      Practice 05 (Tensorflow,Seq2Seq)

      Week 12

      (11/15, 11/17)

      • Recurrent Neural Networks (RNN)
      13   14  

      Week 13

      (11/22, 11/24)

      • Applications of RNN
      Google's NMT  

      Week 14

      (11/29, 12/1)

      • Recent Advancements in Deep Learning
      15   16  

      Week 15

      (12/6, 12/8)

      • Poster Presentation

      Week 16

      (12/13)

      • Final Exam
       

    This page is maintained by Sang-Woo Lee
    Last update: 2016. 12. 26.