publications
publications by categories in reversed chronological order. generated by jekyll-scholar.
2025
-
Amortizing intractable inference in diffusion models for Bayesian inverse problemsIn Proc. Workshop on Machine Learning and the Physical Sciences. Accessed, 2025 -
From discrete-time policies to continuous-time diffusion samplers: Asymptotic equivalences and faster trainingarXiv preprint arXiv:2501.06148, 2025 -
Outsourced diffusion sampling: Efficient posterior inference in latent spaces of generative modelsInternational Conference on Machine Learning (ICML), 2025 -
SEMU: Singular Value Decomposition for Efficient Machine UnlearningInternational Conference on Machine Learning (ICML), 2025 -
Solving Bayesian inverse problems with diffusion priors and off-policy RLIn ICLR 2025 Workshop on Deep Generative Model in Machine Learning: Theory, Principle and Efficacy, 2025 -
Embarrassingly Efficient Unlearning with SVDIn ICML 2025 Workshop on Machine Unlearning for Generative AI, 2025 -
Revisiting the Equivalence of Bayesian Neural Networks and Gaussian Processes: On the Importance of Learning ActivationsIn The 41st Conference on Uncertainty in Artificial Intelligence (UAI), 2025
2024
-
Iterated Denoising Energy Matching for Sampling from Boltzmann DensitiesInternational Conference on Machine Learning (ICML), 2024 -
Amortizing intractable inference in diffusion models for vision, language, and controlAdvances in Neural Information Processing Systems (NeurIPS), 2024 -
Early detection and diagnosis of cancer with interpretable machine learning to uncover cancer-specific DNA methylation patternsBiology Methods and Protocols, 2024 -
Autolora: Autoguidance meets low-rank adaptation for diffusion modelsarXiv preprint arXiv:2410.03941, 2024 -
Improved off-policy training of diffusion samplersAdvances in Neural Information Processing Systems (NeurIPS), 2024 - Hi-fi functional priors by learning activationsIn NeurIPS 2024 Workshop on Bayesian Decision-making and Uncertainty, 2024
2023
-
Hypershot: Few-shot learning by kernel hypernetworksIn Proceedings of the IEEE/CVF winter conference on applications of computer vision, 2023 -
The general framework for few-shot learning by kernel HyperNetworksMachine Vision and Applications, 2023
2021
-
OneFlow: One-class flow for anomaly detection based on a minimal volume regionIEEE Transactions on Pattern Analysis and Machine Intelligence, 2021 -
Flow-based SVDD for anomaly detectionIn ICML Workshop on Invertible Neural Networks, Normalizing Flows, and Explicit Likelihood Models, 2021 -
Missing Glow Phenomenon: Learning Disentangled Representation of Missing DataIn International Conference on Neural Information Processing, 2021 -
Non-gaussian gaussian processes for few-shot regressionAdvances in Neural Information Processing Systems (NeurIPS), 2021
2020
- Supermodeling: the next level of abstraction in the use of data assimilationIn International Conference on Computational Science, 2020
2019
- Data adaptation in handy economy-ideology modelarXiv preprint arXiv:1904.04309, 2019
2018
- Hybrid swarm and agent-based evolutionary optimizationIn International Conference on Computational Science, 2018