I’m an independent researcher fascinated by the intersection of statistics, machine learning, and time series.
My work focuses on blending mathematical rigor with practical modeling, drawing from areas such as high-dimensional statistics, learning theory, probabilistic modeling, generative models, and state-space systems.

I’m especially interested in how theory can guide the design of models that explain, predict, and generate complex data — particularly in real-world, high-dimensional settings.

• Machine Learning Methodologies

  • Time Series: Forecasting, Classification, Representation Learning, Generation
  • Probabilistic & Statistical Machine Learning:
    Learning Theory, Deep Generative Models (Energy-Based, VAE, Flow, Diffusion),
    Approximate Bayesian Inference (MCMC, VI), Uncertainty Quantification
  • High-dimensional Statistics: Variable Selection, Missing Data

• Applications

  • Biostatistics
  • Demand & Sales Forecasting
  • Business Problems: Credit Scoring, Customer Retention, Marketing Mix Modeling,
    Portfolio Optimization, Inventory Optimization

• Programming:
  
  

• Machine Learning Frameworks:
  
  
  JAX

• Probabilistic Programming:
  NumPyro, Pyro, PyMC3

I enjoy exploring connections between mathematics, computation, and application domains — and I’m always curious about how theory and practice can inform each other.