Jianfeng Lu

Professor of Mathematics

Jianfeng Lu is an applied mathematician interested in mathematical analysis and algorithm development for problems from computational physics, theoretical chemistry, materials science and other related fields.

More specifically, his current research focuses include:
Electronic structure and many body problems; quantum molecular dynamics; multiscale modeling and analysis; rare events and sampling techniques.

Appointments and Affiliations

• Professor of Mathematics
• Associate Professor of Chemistry

Contact Information

• Office Location: 242 Physics Bldg, 120 Science Drive, Durham, NC 27708
• Office Phone: (919) 660-2875
• Email Address: jianfeng@math.duke.edu
• Websites:
  • Personal website

Education

• Ph.D. Princeton University, 2009

Awards, Honors, and Distinctions

• IMA Prize in Mathematics and its Applications. Institute of Mathematics and its Applications. 2017
• CAREER Award. National Science Foundation. 2015
• Sloan Research Fellowship. Alfred P. Sloan Foundation. 2013
• Porter Ogden Jacobus Fellowship. Princeton University. 2008

Courses Taught

• MATH 391: Independent Study
• MATH 393: Research Independent Study
• MATH 394: Research Independent Study
• MATH 493: Research Independent Study
• MATH 555: Ordinary Differential Equations
• MATH 631: Measure and Integration
• MATH 660: Numerical Partial Differential Equations
• MATH 690-60: Topics in Numerical Methods

In the News

• Pratt Researchers to Harness Computational Power to Solve Time-Intensive Calculations (Jul 10, 2015)
• Sloan Foundation Names Charbonneau, Lu as 2013 Research Fellows (Feb 15, 2013)

Representative Publications

• Chen, S; Chewi, S; Lee, H; Li, Y; Lu, J; Salim, A, The probability flow ODE is provably fast (2023) [abs].
• Wang, Z; Zhang, Z; Lu, J; Li, Y, Coordinate Descent Full Configuration Interaction for Excited States (2023) [abs].
• Lu, J; Wu, Y; Xiang, Y, Score-based Transport Modeling for Mean-Field Fokker-Planck Equations (2023) [abs].
• An, J; Lu, J, Convergence of stochastic gradient descent under a local Lajasiewicz
  condition for deep neural networks (2023) [abs].
• Wang, M; Lu, J, Neural Network-Based Variational Methods for Solving Quadratic Porous Medium Equations in High Dimensions, Communications in Mathematics and Statistics, vol 11 no. 1 (2023), pp. 21-57 [10.1007/s40304-023-00339-5] [abs].