https://dicai.github.io/project/ Recent content in Projects on Diana Cai Hugo -- gohugo.io en-us Sat, 10 Dec 2022 00:00:00 +0000 https://dicai.github.io/project/biomedical/ Sat, 10 Dec 2022 00:00:00 +0000 https://dicai.github.io/project/biomedical/ With researchers at Princeton, Stanford, and UNC, I am collaborating on several application areas related to the biomedical health sciences: Experimental design for spatial genomics Bayesian inverse reinforcement learning for health applications Efficient online changepoint detection in time series https://dicai.github.io/project/materials/ Sat, 10 Dec 2022 00:00:00 +0000 https://dicai.github.io/project/materials/ Machine learning has tremendous prediction to accelerate the design of novel materials. I currently collaborate with researchers at Princeton, the Colorado School of Mines, UIUC, and WashU on applications of machine learning and probabilistic methods to material science problems. We are developing new methods for active search, Markov Chain Monte Carlo simulation, and combinatorial design. Current application areas include machine learning for alloy design, ML-assisted property prediction, and topology optimization of structural materials. https://dicai.github.io/project/vi/ Sun, 27 Sep 2020 00:00:00 +0000 https://dicai.github.io/project/vi/ Variational inference (VI) approximates an intractable target distribution with the closest member of a tractable family. A central challenge in black-box VI is that standard implementations, which minimize the KL divergence using stochastic gradient descent (SGD), converge slowly due to gradient noise and sensitivity to learning parameters. These issues become more pronounced when using expressive variational families, which are necessary for accurate inference in complex scientific problems. My work shows that score matching—aligning the gradients of the log densities of the target model and variational approximation—enables faster, more reliable optimization, often yielding closed-form or convex subproblems. https://dicai.github.io/project/robust/ Thu, 27 Apr 2017 00:00:00 +0000 https://dicai.github.io/project/robust/ Making probabilistic machine learning more robust to model misspecification https://dicai.github.io/project/inference/ Wed, 27 Apr 2016 00:00:00 +0000 https://dicai.github.io/project/inference/ Accelerating machine learning algorithms for inference, design, and search https://dicai.github.io/project/structured/ Wed, 27 Apr 2016 00:00:00 +0000 https://dicai.github.io/project/structured/ https://dicai.github.io/project/test/ Wed, 27 Apr 2016 00:00:00 +0000 https://dicai.github.io/project/test/ https://dicai.github.io/project/bnp/ Mon, 27 Apr 2015 00:00:00 +0000 https://dicai.github.io/project/bnp/ Developing flexible Bayesian models and studying their properties