Publications

\(^*\) indicates equal contribution

Submitted Papers/In Preparation

W. Song, D. Kim,
Linear loss classification: Efficient training through neural collapse.

J. Chae, D. Kim,
Understaning schedule-free methods in nonconvex optimization: Rate guarantees and escaping saddles.

Y. Kim\(^*\), K. Lee\(^*\) and D. Kim,
Minimax optimization without spurious solutions in optimal transport learning.

S. Lee and D. Kim,
Semi-anchored multi-step gradient descent ascent method for structured nonconvex-nonconcave composite minimax problems.
[arXiv version]

Journal/Conference Papers

K. Lee and D. Kim,
Zero-Infinity GAN: Stable dynamics and implicit bias of extragradient,
Optimization for Machine Learning (OPT) Workshop at Neural Information Processing Systems (NeurIPS), 2025.
- Received the best poster award at the KMS Annual Meeting

J. Chae\(^*\), K. Kim\(^*\) and D. Kim,
Rethinking memorization-generalization trade-off in generative models,
High-dimensional Learning Dynamics (HiLD) Workshop at International Conference on Machine Learning (ICML), 2025.

J. Chae, C. Yun, and D. Kim,
Stochastic extragradient with flip-flop shuffling & anchoring: Provable improvements,
Neural Information Processing Systems (NeurIPS), 2024. [arXiv version]

K. Kim and D. Kim,
Double-step alternating extragradient with increasing timescale separation for finding local minimax points: Provable improvements,
International Conference on Machine Learning (ICML), 2024.

A. Alacaoglu, D. Kim and S. J. Wright,
Revisiting inexact fixed-point iterations for min-max problems: Stochasticity and structured nonconvexity,
International Conference on Machine Learning (ICML), 2024. [arXiv version]

J. Chae, K. Kim and D. Kim,
Two-timescale extragradient for finding local minimax points,
International Conference on Learning Representations (ICLR), 2024. [arXiv version]

J. Chae, K. Kim and D. Kim,
Open Problem: Is there a first-order method that only converges to local minimax optima?,
Conference on Learning Theory (COLT), 2023.

S. Lee and D. Kim,
Fast extra gradient methods for smooth structured nonconvex-nonconcave minimax problems,
Neural Information Processing Systems (NeurIPS), 2021. [arXiv version]

D. Kim,
Accelerated proximal point method for maximally monotone operators,
Mathematical Programming, Series A, vol. 190, no. 1, pp. 57-87, Nov. 2021. [arXiv version]

D. Kim and J. A. Fessler,
Optimizing the efficiency of first-order methods for decreasing the gradient of smooth convex functions,
J. Optim. Theory Appl., vol. 188, no. 1, pp. 192-219, 2021. [arXiv version]

K. Kim, D. Kim, J. Yang, G. El Fakhri, Y. Seo, J. A. Fessler, and Q. Li,
Time of flight PET reconstruction using non-uniform update for regional recovery uniformity,
Medical Physics, vol 46, no. 1, pp. 649-64, Feb. 2019.

D. Kim and J. A. Fessler,
Adaptive restart of the optimized gradient method for convex minimization,
J. Optim. Theory Appl., vol. 178, no. 1, pp. 240-63, Jul. 2018. [arXiv version] [matlab code]

D. Kim and J. A. Fessler,
Generalizing the optimized gradient method for smooth convex minimization,
SIAM J. Optimization, vol. 28, no. 2, pp. 1920-50, 2018. [arXiv version]

D. Kim and J. A. Fessler,
Another look at the fast iterative shrinkage/thresholding algorithm (FISTA),
SIAM J. Optimization, vol. 28, no. 1, pp. 223-50, 2018. [arXiv version]

D. Kim and J. A. Fessler,
Accelerated dual gradient-based methods for total variation image denoising/deblurring problems,
Proc. IEEE Conf. Acoust. Speech Sig. Proc., pp. 6230-4, 2017.

D. Kim and J. A. Fessler,
On the convergence analysis of the optimized gradient method,
J. Optim. Theory Appl., vol. 172, no. 1, pp. 187-205, Jan. 2017. [arXiv version]

D. Kim and J. A. Fessler,
Optimized first-order methods for smooth convex minimization,
Mathematical Programming, Series A, vol. 159, no. 1, pp. 81-107, Sep. 2016. [arXiv version]
- A small error in the published manuscript is corrected in the arXiv version.

D. Kim and J. A. Fessler,
An optimized first-order method for image restoration,
Proc. IEEE Intl. Conf. on Image Proc., pp. 3675-9, 2015.

D. Kim and J. A. Fessler,
Distributed block-separable ordered subsets for helical X-ray CT image reconstruction,
Proc. Intl. Mtg. on Fully 3D Image Recon. in Rad. and Nuc. Med., pp. 138-41, 2015.

D. Kim, S. Ramani, and J. A. Fessler,
Combining ordered subsets and momentum methods for accelerated X-ray CT image reconstruction,
IEEE Trans. Med. Imag., vol. 34, no. 1, pp. 167-78, Jan. 2015.

D. Kim and J. A. Fessler,
Optimized momentum steps for accelerating X-ray CT ordered subsets image reconstruction,
Proc. Third Intl. Mtg. on image formation in X-ray computed tomography, pp. 103-6, 2014.

D. Kim, D. Pal, J.-B. Thibault, and J. A. Fessler,
Accelerating ordered subsets image reconstruction for X-ray CT using spatially non-uniform optimization transfer,
IEEE Trans. Med. Imag., vol. 32, no. 11, pp. 1965-78, Nov. 2013.

D. Kim and J. A. Fessler,
Ordered subsets acceleration with relaxed momentum for X-ray CT image reconstruction,
Proc. IEEE Nuc. Sci. Symp. Med. Im. Conf., 2013.

D. Kim, S. Ramani, and J. A. Fessler,
Accelerating X-ray CT ordered subsets image reconstruction with Nesterov's first-order methods,
Proc. Intl. Mtg. on Fully 3D Image Recon. in Rad. and Nuc. Med., pp. 22-5, 2013.

D. Kim, S. Ramani, and J. A. Fessler,
Ordered subsets with momentum for accelerated X-ray CT image reconstruction,
Proc. IEEE Conf. Acoust. Speech Sig. Proc., pp. 920-3, 2013.

D. Kim, D. Pal, J.-B. Thibault, and J. A. Fessler,
Improved ordered subsets algorithm for 3D X-ray CT image reconstruction,
Proc. Second Intl. Mtg. on image formation in X-ray computed tomography, pp. 378-81, 2012.

D. Kim and J. A. Fessler,
Parallelizable algorithms for X-ray CT image reconstruction with spatially non-uniform updates,
Proc. Second Intl. Mtg. on image formation in X-ray computed tomography, pp. 33-7, 2012.

J. A. Fessler and D. Kim,
Axial block coordinate descent (ABCD) algorithm for X-ray CT image reconstruction,
Proc. Intl. Mtg. on Fully 3D image Recon. in Rad. and Nuc. Med., pp. 262-5, 2011.

D. Kim and J. A. Fessler,
Accelerated ordered-subsets algorithm based on separable quadratic surrogates for regularized image reconstruction in X-ray CT,
Proc. IEEE Intl. Symp. Biomed. Imag., pp. 1132-7, 2011.

Technical reports

D. Kim and J. A. Fessler,
Fast dual proximal gradient algorithms with rate \(O(1/k^{1.5})\) for convex minimization.
[arXiv version]