Shopping Carts — optimized with #DeepLearning, Genetic Algorithms, and GPUs
The problem — when multiple vendors (each with multiple warehouses!) can fulfill the same item, what combination of vendirs/warehouses minimizes the total cost of the shopping cart- inclusive of price, shipping, and commissions.And remember, the more items in the cart, the more expensive the search, but the greater the savings too.
The solution —
- • Genetic algorithms for searching
- • #MachineLearning to find optimal algorithms
- • CUDA to run these algorithms
Read on for the fascinating details — in particular, how AI/ML looks like it will rejuvenate genetic algos, the intersection of monster search spaces and complex iterative feedback systems seem to be a natural fit. (°)
Mind you, the ability to naturally parallelize the process doesn’t hurt
(°) Just one example — should the GPU be used? Choosing the best algorithm for the task is unfortunately not as simple as “if cart = small then do CPUFS elif cart = big then do GPUFS elif cart = huge then do GPUGA.” We needed a smart decision function to address these issues so we used machine learning to train a model based on cart features and used it to improve the function’s results
Comments