https://www.math.snu.ac.kr/board/files/attach/images/701/ff97c54e6e21a4ae39315f9a12b27314.png
Extra Form
강연자 신연종
소속 KAIST
date 2022-10-13

 

Machine learning (ML) has achieved unprecedented empirical success in diverse applications. It now has been applied to solve scientific problems, which has become an emerging field, Scientific Machine Learning (SciML). Many ML techniques, however, are very complex and sophisticated, commonly requiring many trial-and-error and tricks. These result in a lack of robustness and interpretability, which are critical factors for scientific applications. This talk centers around mathematical approaches for SciML, promoting trustworthiness. The first part is about how to embed physics into neural networks (NNs). I will present a general framework for designing NNs that obey the first and second laws of thermodynamics. The framework not only provides flexible ways of leveraging available physics information but also results in expressive NN architectures. The second part is about the training of NNs, one of the biggest challenges in ML. I will present an efficient training method for NNs - Active Neuron Least Squares (ANLS). ANLS is developed from the insight gained from the analysis of gradient descent training.

Atachment
첨부 '1'
  1. Birational Geometry of varieties with effective anti-canonical divisors

    Fano varieties are fundamental objects in algebraic geometry. These can be considered as the unique output of the -K -minimal model program on the varieties with effective anticanonical divisors. Thus the initial models should encode the in...
    Category수학강연회 소속연세대학교 강연자최성락
    Read More
  2. Contact instantons and entanglement of Legendrian links

    We introduce a conformally invariant nonlinear sigma model on the bulk of contact manifolds with boundary condition on the Legendrian links in any odd dimension. We call any finite energy solution a contact instanton. We also explain its Ha...
    Category수학강연회 소속IBS-CGP /POSTECH 강연자오용근
    Read More
  3. <학부생을 위한 ɛ 강연> Self-Supervised Learning in Computer Vision

    In recent years, artificial intelligence has made remarkable progress in developing algorithms that can learn from vast amounts of carefully labeled data. This paradigm of supervised learning has made great success in training specialist mo...
    Category수학강연회 소속인하대학교 강연자현윤석
    Read More
  4. <학부생을 위한 강연> 수학과 보험산업

    Category수학강연회 소속라이나생명 강연자유신옥
    Read More
  5. Counting number fields and its applications

    It is a fascinating and challenging problem to count number fields with bounded discriminant. It has so many applications in number theory. We give two examples. First, we compute the average of the smallest primes belonging to a conjugacy ...
    Category수학강연회 소속UNIST 강연자조재현
    Read More
  6. 2022-2 Rookies Pitch: Algebraic Geometry (박현준)

    CategoryBK21 FOUR Rookies Pitch 소속KIAS 강연자박현준
    Read More
  7. 2022-2 Rookies Pitch: Geometric Topology (김승원)

    CategoryBK21 FOUR Rookies Pitch 소속성균관대학교 강연자김승원
    Read More
  8. 2022-2 Rookies Pitch: Probability Theory (이중경)

    CategoryBK21 FOUR Rookies Pitch 소속수리과학부 강연자이중경
    Read More
  9. 2022-2 Rookies Pitch: Probability Theory (변성수)

    CategoryBK21 FOUR Rookies Pitch 소속KIAS 강연자변성수
    Read More
  10. Towards Trustworthy Scientific Machine Learning: Theory, Algorithms, and Applications

    Machine learning (ML) has achieved unprecedented empirical success in diverse applications. It now has been applied to solve scientific problems, which has become an emerging field, Scientific Machine Learning (SciML). Many ML techniques, h...
    Category수학강연회 소속KAIST 강연자신연종
    Read More
  11. Elliptic equations with singular drifts in critical spaces

    Category수학강연회 소속서강대학교 강연자김현석
    Read More
  12. Contact topology of singularities and symplectic fillings

    For an isolated singularity, the intersection with a small sphere forms a smooth manifold, called the link of a singularity. It admits a canonical contact structure, and this turns out to be a fine invariant of singularities and provides an...
    Category수학강연회 소속순천대학교 강연자권명기
    Read More
  13. 2022-2 Rookies Pitch: Representation Theory(이신명)

    CategoryBK21 FOUR Rookies Pitch 소속수리과학부 강연자이신명
    Read More
  14. 2022-2 Rookies Pitch: Representation Theory(허태혁)

    CategoryBK21 FOUR Rookies Pitch 소속QSMS 강연자허태혁
    Read More
  15. 2022-2 Rookies Pitch: Harmonic Analysis (오세욱)

    CategoryBK21 FOUR Rookies Pitch 소속고등과학원 강연자오세욱
    Read More
  16. 2022-2 Rookies Pitch: Harmonic Analysis (함세헌)

    CategoryBK21 FOUR Rookies Pitch 소속수학연구소 강연자함세헌
    Read More
  17. Combinatorics and Hodge theory

    I will tell two interrelated stories illustrating fruitful interactions between combinatorics and Hodge theory. The first is that of Lorentzian polynomials, based on my joint work with Petter Brändén. They link continuous convex...
    Category특별강연 소속미국 프린스턴대 교수, 한국 고등과학원 석학교수 강연자허준이
    Read More
  18. 2022-1 Rookies Pitch: Harmonic Analysis (Kalachand Shuin)

    CategoryBK21 FOUR Rookies Pitch 소속BK21 강연자Kalachand Shuin
    Read More
  19. 2022-1 Rookies Pitch:Functional Analysis (정민구)

    CategoryBK21 FOUR Rookies Pitch 소속고등과학원 강연자정민구
    Read More
  20. Regularity theory for non-autonomous elliptic equations in divergence form

    Category수학강연회 소속서강대학교 강연자옥지훈
    Read More
Board Pagination Prev 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Next
/ 14