Bayesian Inference + DL (TensorFlow, PyMC3, FEniCS)

Ongoing work on improving reduced order models by better characterizing their error using deep learning methods.

---

Deep Learning Surrogates (TensorFlow, FEniCS)

Deep neural network surrogate models to solve inverse heat conduction problems.

---

Inverse problems in two-photon acoustic tomography (FEniCS)

A limited-memory BFGS optimization routine to compute maximum a posteriori estimates given observations to reconstruct parameters in photo-acoustic tomography modeled by semi-linear diffusion equations.

---

Parallel particle mesh method in N-body problems (FFTW)

Parallel discrete Fourier transform for scalable N-body simulations

---

Personal Website

A simple website based on Jekyll theme: minima.