The University of Osaka
Machine Learning & Systems Laboratory

This is a yearly list of major publications since October 2022 when the laboratory at Osaka Univ. was established.
For the previous publications, please see each faculty member's page Member ».

2026〜

Refereed Papers

• K. Hayashi, A. Yamamoto, and Y. Kawahara, "Reduced-Order Smoke Animation with Immersed-Boundary Forcing for Moving Rigid Bodies," in Proc. of the 19th ACM SIGGRAPH Conf. and Exhibition on Computer Graphics and Interactive Techniques in Asia (SIGGRAPH Asia 2026) (Posters) (accepted).
• H. Kawai, R. Tanaka, T. Konishi, and Y. Kawahara, "Deep Operator World Models for Policy Mirror Descent in Reinforcement Learning," in Proc. of the 33rd Int'l Conf. on Neural Information Processing (ICONIP'26) (accepted).
• M. Mitani, S. Noguchi, T. Konishi, and Y. Kawahara, "Wave-Dynamics-Augmented Graph Neural Networks for Mesh-Based Physical Simulation," in Proc. of the 33rd Int'l Conf. on Neural Information Processing (ICONIP'26) (accepted).
• K. Yamazaki, I. Sakata, T. Konishi, and Y. Kawahara, "Physics-inspired transformer quantum states via latent imaginary-time evolution," Physical Review Research, Vol.8, Article Number 033032, 2026 (the arXiv version is found [Here]).
• N. Ke, P. Thanasutives, and Y. Kawahara, "Structured Noise Adaptation for Sequential Bayesian Filtering with Embedded Latent Transfer Operators," Trans. on Machine Learning Research, ISSN 2835-8856, 2026.
• R. Tanaka and Y. Kawahara, "Deep Spectral Learning of Embedded Latent Transfer Operators for Stochastic Dynamical Systems," in Proc. of the 42nd Conf. on Uncertainty in Artificial Intelligence (UAI'26), PMLR 337:6610-6630, 2026.
• K. Imamura, A. Nagahashi, A. Okusa, T. Yamamoto, Y. Izumi, N. Ueda, Y. Kawahara, and H. Inoue, "Non-linear combinatorial analysis of blood transcriptomes identifies PRKAR1A as a regulator of TDP-43 pathology in amyotrophic lateral sclerosis," Biology Methods & Protocols, Vol.11, Issue 1, bpag023, 2026.
• S. Noguchi and Y. Kawahara, "Mesh Field Theory: Port-Hamiltonian Formulation of Mesh-Based Physics," in Proc. of the 43rd Int'l Conf. on Machine Learning (ICML'26) (accepted) (the arXiv version is found [Here]).
• K. Fujii, K. Tsutsui, Y. Teshima, M. Itoh, N. Takeishi, N. Nishiumi, R. Tanaka, and Y. Kawahara, "Data-driven simulator of multi-animal behavior with unknown dynamics via reinforcement learning," iScience, Vol.29, No.6, Article Number 116192, 2026.
• Y. Yamamoto, S. Isono, Y. Kawahara, and S. Yokoi, "Timesteps of Mamba Align with Human Reading Times," in Proc. of the 64th Annual Meeting of the Association for Computational Linguistics (ACL'26) (findings), pp.31820–31832, 2026.
• Y. Inui, M. Ikeda, T. Konishi, and Y. Kawahara, "Kernel Occupation Readout for Oscillatory Recurrent Neural Networks," in Proc. of the Int'l Joint Conf. on Neural Networks (IJCNN'26) (accepted).
• M. Takashima, T. Horike, K. Takahashi, M. Yamada, K. Iwakiri, H. Miyamura, R. Takeuchi, and Y. Kawahara, "Development of an Optimization System of Rotor Profile for Single Screw Compressors Using Evolutionary Computation," in Proc. of the 28th Int'l Compressor Engineering Conf. (in 2026 Herrick Conferences), Paper ID 1300, 2026.
• T. Yamada, T. Konishi, and Y. Kawahara, "Predictive Fair Representation Learning with Variational Autoencoders," New Generation Computing, Vol.44, Article Number 16, 2026.
• U. Arshad, A. Tubaisha, S. Anwar, Z. Halim, and Y. Kawahara, "Blockchain Enabled EEG Neuromarketing Framework for Secure and Intelligent Consumer Electronics," IEEE Trans. on Consumer Electronics, Vol.72, No.3, pp.6796-6803, 2026.
• M. Wadinger, M. Kvasnica, and Y. Kawahara, "Truncated Online Dynamic Mode Decomposition with Control for Industrial Change Detection," in Proc. of the 24th European Control Conference (ECC'26), pp.797-802, 2026.
• Y. Wu and Y. Kawahara, "An Operator Analysis on Stochastic Differential Equations (SDEs)-Based Diffusion Generative Models," Entropy, Vol.28, No.3, Article Number 290, 2026.
• K. Yamazaki, T. Konishi, and Y. Kawahara, "Explorative Curriculum Learning for Strongly Correlated Electron Systems," ACM Trans. on AI for Science (ACM TAIS) (accepted) (the arXiv version is found [Here]).
• J. Bai and Y. Kawahara, "Dualformer: Time-Frequency Dual Domain Learning for Long-term Time Series Forecasting," in Proc. of the 29th Int'l Conf. on Artificial Intelligence and Statistics (AISTATS'26) (accepted) (the arXiv version is found [Here]).
• Y. Hashimoto, S. Sonoda, I. Ishikawa, and M. Ikeda, "Why High-rank Neural Networks Generalize?: An Algebraic Framework with RKHSs," in Proc. of the 14th Int'l Conf. on Learning Representations (ICLR'26) (accepted).
• Y. Miyauchi, M. Ikeda, and Y. Kawahara, "Generalized Stochastic Resilience for Early Warning Signals Based on Koopman Operator," Nonlinear Dynamics, Vol. 114, Article Number 246, 2026 (the arXiv version is found [Here]).
• X. Yang, P. Qiu, and Y. Kawahara, "Infinitely deep Bayesian neural network with signature transform," Neurocomputing, Vol. 670, Paper Number 132563, 2026.

Book Chapter

• H. Arai, Y. Kawahara, T. LeToan, K. Oyoshi, W. Takeuchi, S.V. Nghiem, M. Zribi, T. Fumoto, K. Inubushi, L.D. Nguyen, S. Sobue, and T. Ohara, "Global land-Atmospheric Data assimilation based on Koopman operator error covariance in a deep neural network latent space for GHG emission verification," in Agriculture, Forestry and Other Land Use (AFOLU) Emissions in South and Southeast Asia (eds. Krishna Vadrevu, Toshimasa Ohara, and Chris Justice), Springer (to appear).

2025

Refereed Papers

• I. Ishikawa, Y. Hashimoto, M. Ikeda, and Y. Kawahara, "Koopman operators with intrinsic observables in rigged reproducing Kernel Hilbert spaces," Nonlinearity, Vol. 38, No.11, Article Number 115022, 2025 (the arXiv version is found [Here]).
• Kimihiro Yamazaki, Itsushi Sakata, Takuya Konishi, and Yoshinobu Kawahara, "Weight-sharing Transformer quantum states with Suzuki–Trotter decompositions," Machine Learning and the Physical Sciences Workshop at the 39th conference on Neural Information Processing Systems (NeurIPS), December. 2025.
• S. Noguchi and Y. Kawahara, "Wavy Transformer," Advances in Neural Information Processing Systems (NeurIPS'25), 2025 (the arXiv version is found [Here]).
• †M. Fujisawa, †M. Adachi, and M. A. Osborne (†:Equal contribution), "Scalable Valuation of Human Feedback through Provably Robust Model Alignment," Advances in Neural Information Processing Systems (NeurIPS'25), 2025.
• †F. Futami and †M. Fujisawa (†:Equal contribution), "Information-theoretic Generalization Analysis for VQ-VAEs: A Role of Latent Variables," Advances in Neural Information Processing Systems (NeurIPS'25), 2025.
• P. Thanasutives, Y. Kawahara, and K. Fukui "Bayesian Model Selection for Parametric Variable-coefficient Differential Equation Discovery," Results in Engineering, 106930, 2025.
• †M. Fujisawa and †F. Futami (†equal contribution), "On the Convergence of SVGD in KL divergence via Approximate gradient flow," Trans. on Machine Learning Research, 2025.
• B.J. Kim, Y. Kawahara, and S.W. Kim, "Disappearance of Timestep Embedding: A Case Study on Neural ODE and Diffusion Models," Trans. on Machine Learning Research, 2025.
• M. Adachi, M. Fujisawa, and M. A. Osborne, "Fixing the Pitfalls of Probabilistic Time-Series Forecasting Evaluation by Kernel Quadrature," in Proc. of the First Int'l Conf. on Probabilistic Numerics (ProbNum25) (accepted).
• †M. Fujisawa and †F. Futami (†:Equal contribution), "PAC-Bayes Analysis for Recalibration in Classification," in Proc. of the 42nd Int'l Conf. on Machine Learning (ICML'25) (accepted).
• S. Sonoda, Y. Hashimoto, I. Ishikawa, and M. Ikeda, "Unified universality theorem for deep and shallow joint-group equivariant machines," in Proc. of the 42nd Int'l Conf. on Machine Learning (ICML'25) (accepted)
• †N. Ke, †R. Tanaka and Y. Kawahara (†equal contribution), "Learning Stochastic Nonlinear Dynamics with Embedded Latent Transfer Operators," in Proc. of the 28th Int'l Conf. on Artificial Intelligence and Statistics (AISTATS'25), PMLR 258:4861–4869, 2025 (the arXiv version is found [Here]).
• K.Ghalamkari, J.L. Hinrich, and M. Mørup, "Non-negative Tensor Low-rank Decompositions Through the Lens of Information Geometry," AAAI 25 Workshop on Connecting Low-rank Representations in AI, March. 2025.
• K. Fujii, K. Takeuchi, A. Kuribayashi, N. Takeishi, Y. Kawahara, and K. Takeda, "Estimating counterfactual treatment outcomes over time in complex multi-agent scenarios," IEEE Transactions on Neural Networks and Learning Systems, Vol.36, No.2, pp.2103-2117, 2025.

2024

• M. Weissenbacher, E. Routis and Y. Kawahara, "Self-supervised Color Generalization in Reinforcement Learning," Transactions on Machine Learning Research (TMLR), 2024.
• Y. Inui, T. Konishi, and Y. Kawahara, "Learning with Almost Invariant Sets in Neural Oscillatory ODEs," in Proc. of the 31st Int'l Conf. on Neural Information Processing (ICONIP'24) (accepted).
• I. Sakata and Y. Kawahara, "Enhancing Spectral Analysis in Nonlinear Dynamics: An Approach Based on Pseudoeigenfunctions from Continuous Spectra," Scientific Reports, Vol.14, Article Number 19276, 2024.
• M. Ohnishi, I. Ishikawa, K. Lowrey, M. Ikeda, S.M. Kakade, and Y. Kawahara, "Koopman Spectrum Nonlinear Regulators and Efficient Online Learning," Transactions on Machine Learning Research (TMLR), 2024.
• R. Fukuma, K. Majima, Y. Kawahara, O. Yamashita, Y. Shiraishia, H. Kishima, and T. Yanagisawa, "Fast, accurate, and interpretable decoding of electrocorticographic signals using dynamic mode decomposition," Communications Biology, Vol.7, Article number 595, 2024.
• M. Weissenbacher, R. Agarwal, and Y. Kawahara, "SiT: Symmetry-invariant Transformers for Generalisation in Reinforcement Learning," in Proc. of the 41st Int'l Conf. on Machine Learning (ICML'24), PMLR 235:52695-52719, 2024.
• J. Bai and Y. Kawahara, ``MANet: Mixed Attention Network for Visual Explanation," New Generation Computing, DOI: 10.1007/s00354-024-00252-7, 2024.
• K. Fujii, K. Takeuchi, A. Kuribayashi, N. Takeishi, Y. Kawahara, and K. Takeda, "Estimating counterfactual treatment outcomes over time in complex multi-agent scenarios," IEEE Transactions on Neural Networks and Learning Systems (accepted).
• H. Endo, S. Ikeda, K. Harada, H. Yamagata, T. Matsubara, K. Matsuo, Y. Kawahara, and O. Yamashita, "Manifold alteration between major depressive disorder and healthy control subjects using dynamic mode decomposition in resting-state fMRI data," Frontiers in Psychiatry, Vol.15, Article Number 1288808, 2024.
• K. Taga, Y. Kato, Y. Yamazaki, Y. Kawahara, and H. Nakao, "Dynamic mode decomposition for Koopman spectral analysis of elementary cellular automata," Chaos: An Interdisciplinary Journal of Nonlinear Science, Vol.34, No.1, Article Number 013125, 2024 (the arXiv version is found [Here]).
• K. Fujii, K. Tsutsui, A. Scott, H. Nakahara, N. Takeishi, and Y. Kawahara, "Adaptive action supervision in reinforcement learning from real-world multi-agent demonstrations," in Proc. of the 16th Int'l Conf. on Agents and Artificial Intelligence (ICAART'24), Volume 2: ICAART, 27-39 (the arXiv version is found [Here]).
• K. Fujii, N. Takeishi, Y. Kawahara, and K. Takeda, "Decentralized Policy Learning with Partial Observation and Mechanical Constraints for Multi-person Modeling Learning Systems," Neural Networks, Vol.171, pp.40-52, 2024.

2023

• K. Ghalamkari, M. Sugiyama, and Y. Kawahara, "Many-body Approximation for Non-negative Tensors," Advances in Neural Information Processing Systems 36 (NeurIPS'23) (the arXiv version is found [Here]).
• N. Takeishi and Y. Kawahara, "A Characteristic Function for Shapley-Value-Based Attribution of Anomaly Scores," Transactions on Machine Learning Research (TMLR), 2023.
• D. Bodenham and Y. Kawahara, "euMMD: Efficiently Computing the MMD Two-Sample Test Statistic for Univariate Data," Statistics and Computing, Vol.33, Article Number 110, 2023.
• T. Konishi and Y. Kawahara, "Stable Invariant Models with Koopman Spectra," Neural Networks, Vol.165, pp.393-405, 2023 (the arXiv version is found [Here]).
• T. Iwata and Y. Kawahara, "Neural dynamic mode decomposition for end-to-end modeling of nonlinear dynamics," Journal of Computational Dynamics, Vol.10, No.2, pp.268-280, 2023.

2022

(Since October 2022)

• N. Takeishi, K. Takeuchi, K. Fujii, and Y. Kawahara, "Discriminant Dynamic Mode Decomposition for Labeled Spatio-Temporal Data Collections," SIAM Journal on Applied Dynamical Systems, Vol.21, No.2, pp.1030–1058, 2022.
• H. Arai, M. Zribi, K. Oyoshi, K. Dassas, M. Huc, S. Sobue and T. Le Toan, "Quality Control of CyGNSS Reflectivity for Robust Spatiotemporal Detection of Tropical Wetlands," Remote Sensing, Vol.14, No.22, Article No.:5903, 2022.
• K. Fujii, K. Takeuchi, A. Kuribayashi, N. Takeishi, Y. Kawahara, and K. Takeda, "Estimating counterfactual treatment outcomes over time in multi-vehicle simulation," in Proc. Of the 30th ACM SIGPATIAL Int'l Conf. on Advances in Geographic Information Systems (SIGSPATIAL'22) (short paper), Article No.: 7, pp.1-4, 2022 (This paper received the Best Poster Award).