We are constrained by time in this course, and there is a lot more to learn about in terms of optimisation methods. This page lists many references to important keywords in this field.
- Learn how to solve quadratic problems with SLSQP;
- Learn about quasi-Newton methods for fast gradient converging methods like BFGS;
- Learn about constrained optimisation with the Karush-Kuhn-Tucker (KKT) theorem.
- Learn about common optimisation methods for ML
- Learn about interior-point methods;
- Learn about the duality theorem
- Learn about common ways to linearise constraints;
- Learn about more advanced cuts, e.g. Gomory cuts;
- Learn about advanced modelling tricks, like Benders' decomposition or Dantzig-Wolfe decomposition
- Learn about polynomial reductions;
- Learn about complexity categories above NP
- Learn about more propagation mechanisms for global constraints;
- Learn about local search and it fits CP for large-neighbourhood search;
- Learn about SAT and SMT theories
- Learn about more metaheuristics such as Particle Swarm Optimisation, Ant Colony Optimisation, and more;
- Learn about genetic programming, and see how we can make programs, graphs and networks evolve;
- Learn how evolutionary methods can help optimise the training of neural networks hyperparameters