Coursework completed for Mathematical Inverse Methods in Earth and Environmental Sciences -- Integrated Data Analytics II. Focus on data-driven modeling to solve inverse problems and estimate model parameters. The course was taught by Dr. Suzan van der Lee in Winter 2021.
The final project, completed with Caroline Webster and Eli Ganz, was an extension of coursework. We applied the Monte Carlo, Genetic Algorithm, Simulated Annealing and Markov-Chain Monte Carlo Methods to strongly nonlinear inverse problems.
Figure above: (Left) Graph of objective function contours and solutions from each method in log space. All solutions lie within the center of the local minima. (Right) Graph of objective function contours and solutions from each method in linear space. All solutions lie within the center of the local minima.