TriOptima are currently looking for two, ambitious master thesis students for two separate theses.
They strongly believe that a master thesis is for the benefit of the student so they are flexible when
it comes to formulating a suitable topic in mutual interest.
They are also looking for a financial developer.
For further information, please see: https://careers.
Suggestions for the master thesis in their preferred domain are:
“Determine for which matrix dimensions it is possible to achieve a speedup using a GPU instead of a CPU when doing a Householder reflection-based rank-revealing QR factorization”
“Determine CMA-ES convergence properties for different orthogonal nullspace bases when solving optimization problems with constraints Ax=0”
“Modeling IM calculations using neural network techniques”
“Optimizing SIMM IM using Automatic Differentiation and Gradient Descent”
“Determine which GPU memory allocator strategy that is most beneficial for a certain access pattern and matrix size”
“Compare GPU and CPU Monte Carlo simulations for pricing of derivatives and determine when it is beneficial to use each architecture”
“Compare the Libor Market Model to Stochastic Drift using the Probability Matrix Method for pricing of exotic interest rate derivatives”