research

My main research interests include:

  • causal inference – particularly causal structure learning and problems with confounders
  • Fusion methods – particularly Bayesian model combination
  • Bayesian optimization – particularly in dynamic settings and with non-conventional surrogate models

publications

2024

  1. A Gaussian Process-based Streaming Algorithm for Prediction of Time Series With Regimes and Outliers
    Daniel Waxman, and Petar M. Djurić
    2024
    Submitted
  2. TMLR
    Dynamic Online Ensembles of Basis Expansions
    Daniel Waxman, and Petar M. Djurić
    Transactions on Machine Learning Research (TMLR), 2024
    Accepted.
  3. OJ-SP
    DAGMA-DCE: Interpretable, Non-Parametric Differentiable Causal Discovery
    Daniel WaxmanKurt Butler, and Petar M. Djurić
    IEEE Open Journal of Signal Processing, 2024

2023

  1. ACSSC ’23
    Fusion of Gaussian Process Predictions With Monte Carlo
    Marzieh Ajirak, Daniel Waxman, Fernando Llorente, and Petar M. Djurić
    In 2023 57th Asilomar Conference on Signals, Systems, and Computers, 2023
  2. EUSIPCO ’23
    Detecting Confounders in Multivariate Time Series Using Strength of Causation
    Yuhao Liu, Chen Cui, Daniel WaxmanKurt Butler, and Petar M. Djurić
    In 2023 31st European Signal Processing Conference (EUSIPCO), 2023

talks

  • “Causal Inference via Quantifying Influences” at the Acoustics Research Institute of the Austrian Academy of Sciences (Institut für Schallforschung der Österreichische Akademie der Wissenschaften) [abstract link] [slides]
  • “Bayesian Combination” at the 2023 Bellairs Workshop on Machine Learning and Statistical Signal Processing for Data on Graphs