I’m an independent researcher fascinated by the intersection of statistics, machine learning, and time series. My work blends mathematical rigor with practical modeling, drawing from areas such as high-dimensional statistics, learning theory, probabilistic modeling, generative models, and state-space systems.
- Machine learning methodologies:
- Time Series (TS): TS Forecasting, TS Classification, TS Representation Learning, TS Generation.
- Probabilistic/Statistical Machine Learning: Learning Theory, Deep Generative Models (Energy-Based Models, VAE, Flow Models, Diffusion Models), Approximate Bayesian Inference (MCMC, VI), Uncertainty Quantification.
- High-dimensional Statistics: Variable Selection, Missing Data.
- Real-world applications: High-dimensional Problems in Biostatistics, Demand/Sales Forecasting, Business Problems (Credit Scoring, Customer Retention,Marketing Mixed Modelling, Portfolio Optimization, Inventory Optimization).
- Programming languages:
,
.
- Machine learning (Deep learning) frameworks:
,
, JAX.
- Probabilistic Programming: NumPyro, Pyro, PyMC3.
