Conjugate Gradient method using 1D golden search method to solve minimisation problems + Extension to include constraints & penalty functions
Budget: £20 – £250 GBP
Implement the conjugate gradient method in either
Matlab or Python. Find the most appropriate value for
this parameter using a one-dimensional optimization routine (Golden search). For the β parameter use
Polak-Riviere form or Fletcher-Reeves. Initial objective function will be provided.
The second part is to create a non-linear optimisation problem which includes penalty functions or constraints (that can be solved using the code described above) or implementing the subgradient Nesterov methods. Proof of concept/application of this problem required.
Matlab or Python. Find the most appropriate value for
this parameter using a one-dimensional optimization routine (Golden search). For the β parameter use
Polak-Riviere form or Fletcher-Reeves. Initial objective function will be provided.
The second part is to create a non-linear optimisation problem which includes penalty functions or constraints (that can be solved using the code described above) or implementing the subgradient Nesterov methods. Proof of concept/application of this problem required.