11¿ù 23ÀÏ(¸ñ) <¿Â¶óÀÎ °³ÃÖ> |
³»¿ë |
½Ã°£ |
¼¼ºÎ³»¿ë |
Æ©Å丮¾ó ¼¼¼Ç 1 |
13:00-15:00 |
Robot Learning / ÀÓÀçȯ ±³¼ö(KAIST) |
3D-aware Image Generation and Editing Approaches and Applications / ÁÖÀç°É ±³¼ö(KAIST) |
Fairness in Machine Learning: Challenges, Analysis, and Solutions / ¼Û°æ¿ì ±³¼ö(¿¬¼¼´ë) |
ÈÞ½Ä |
15:00-15:20 |
ÈÞ½Ä |
Æ©Å丮¾ó ¼¼¼Ç 2 |
15:20-17:20
* 16:00-18:00
|
Recent Trends of Distillation Methods in Diffusion Models / ±èµ¿ÁØ ¹Ú»ç(Stanford)
|
Injecting output constraints and dependencies into neural models / ÀÌÀçÀ± ±³¼ö(¼¿ï´ë) |
* Recent Advances in Language-driven Computer Vision Tasks / ±è½Â·æ ±³¼ö(°í·Á´ë) |
11¿ù 24ÀÏ(±Ý) Àå¼Ò : ³×À̹ö 1784 28Ãþ ½ºÄ«ÀÌȦ |
³»¿ë |
½Ã°£ |
¼¼ºÎ³»¿ë |
°³È¸¼±¾ð |
09:00-09:10 |
°³È¸¼±¾ð/ ±è¿ë´ë ȸÀå Ãà»ç/ ÀÌÇö±Ô PM(IITP) |
Special invited talk 1 |
09:10-10:00 |
Optimal Transport in Large-Scale Machine Learning Applications/ Prof. Nhat Ho(University of Texas at Austin) |
±¹Á¦ÇÐȸ¿ì¼ö³í¹®¹ßÇ¥ |
10:00-10:20 |
Universal Few-shot Learning of Dense Prediction Tasks with Visual Token Matching, ICLR 2023/ È«½ÂÈÆ ±³¼ö(KAIST) |
10:20-10:40 |
H-Likelihood Approach to Deep Neural Networks with Temporal-Spatial Random Effects for High-Cardinality Categorical Features, ICML 2023/ ÀÌÇ׺ó ¹Ú»ç(¼¿ï´ë), ÀÌ¿µÁ¶ ±³¼ö(¼¿ï´ë) |
ÈÞ½Ä |
10:40-11:00 |
ÈÞ½Ä |
¿ì¼ö³í¹® ½Ã»ó ¹× ÃÖ¿ì¼ö³í¹® ¹ßÇ¥ |
11:00-11:30 |
¿ì¼ö³í¹® ½Ã»ó ¹× ÃÖ¿ì¼ö³í¹® ¹ßÇ¥ |
Æ÷½ºÅÍ |
11:30-12:00 |
Æ÷½ºÅÍ ¹ßÇ¥ |
Á¡½É |
12:00-13:00 |
Á¡½É |
Plenary Talk |
13:00-13:50 |
Martingale Posterior Distributions/ Prof. Stephen G. Walker(University of Texas at Austin) |
±âȹ¼¼¼Ç 1 |
13:50-14:50 |
ÀΰøÁö´É ÀÌ·Ð |
Cascading Contextual Assortment Bandits/ ¿À¹Îȯ ±³¼ö(¼¿ï´ë) |
Score-based Generative Models in Hilbert Spaces/ ÀÓ¼ººó ±³¼ö(°í·Á´ë) |
Practical Sharpness-Aware Minimization Cannot Converge All the Way to Optima/ À±Ã¶Èñ ±³¼ö(KAIST) |
ÈÞ½Ä |
14:50-15:05 |
ÈÞ½Ä |
±âȹ¼¼¼Ç 2 |
15:05-16:05 |
ºñÁ¯ |
On the Training-Free Image Manipulation using Diffusion Probabilistic Models/ ÇѺ¸Çü ±³¼ö(¼¿ï´ë) |
Multimodal Foundation Models for Video QA/ ±èÇö¿ì ±³¼ö(°í·Á´ë) |
±âȹ¼¼¼Ç 3 |
16:05-17:05 |
ÀÚ¿¬¾î ¹× À½¼º |
X-SNS: Cross-Lingual Transfer Prediction through Sub-Network Similarity/ ±èÅÂ¿í ±³¼ö(ÇѾç´ë) |
ÃֽŠÀ½¼ºÀÎ½Ä ¹× À½¼ºÇÕ¼º °¡¼ú¸®ºä/ ÀåÁØÇõ ±³¼ö(ÇѾç´ë) |
ÈÞ½Ä |
17:05-17:20 |
ÈÞ½Ä |
Special invited talk 2 |
17:20-18:10 |
Controllable AI: Lessons from the Past & The Future of Conversational AI/ ÃÖÁøÈ£ ±³¼ö(Emory University) |
ÃÊû°¿¬ |
18:10-18:40 |
Hyperscale AI, ¿ì¸®ÀÇ °æÀï·Â/ ¼º³«È£ ÇÏÀÌÆÛ½ºÄÉÀÏ AI ±â¼úÃÑ°ý(³×À̹öŬ¶ó¿ìµå) |
¸¸Âù |
19:20 |
¸¸Âù <¶ó¿Â½ºÄù¾î> |
11¿ù 25ÀÏ (Åä) Private Session <³»ºÎ ¿öÅ©¼ó> |