AI Optimization Methods for Data Science
MP4 | Video: AVC 1280x720 | Audio: AAC 44KHz 2ch | Duration: 1 Hour | 90 MB
Genre: eLearning | Language: English

Optimization Methods Overview. This first video in the series covers the rationale for optimization methods along with an overview to the main optimization tools, including deterministic, stochastic, and hybrid systems.
Particle Swarm Optimization Method. This second video in the series covers the Particle Swarm Optimization (PSO) Method, which is one of the most fundamental optimizers. Learn how this heuristic algorithm can approximate global optimum similar to a swarm of bees. PSO pseudocode is provided, along with available package in Julia and Python. Both the strengths and weaknesses are covered, along with its most common use cases.
Genetic Algorithms. This third video in the series covers the genetic optimization algorithm framework, which is one of the most popular optimization methods. Genetic Algorithms (GAs) work as a group of cells that evolve over many generations. GA pseudocode is provided, along with available package in Julia and Python. Both the strengths and weaknesses are covered, along with its most common use cases.
Simulated Annealing Method. This fourth video in the series covers the Simulated Annealing (SA), a heuristic algorithm that works similar to the process of a metal cooling. SA pseudocode is provided, along with available package in Julia and Python. Both the strengths and weaknesses are covered, along with its most common use cases.
Optimization Ensembles. This fifth video in the series covers Optimization Ensembles, including the concept of variants with add-ons or differences from other optimization methods with the goal of improving performance. Some ensemble methods run PSOs, GAs, and SAs in parallel opting to solve the same program. Method SA pseudocode is provided, along with available package in Julia and Python. Both the strengths and weaknesses are covered, along with its most common use cases.