Fun with neural networks
I don’t really feel like writing about demand forecasting. Today is a day of fun.
We talk a lot with our customers about the cool things that neural networks can do within our product, but what kind of things can neural networks do just for fun? Check this out…
Artist Denis Shiryaev uses the facial expressions of people from YouTube and TikTok with neural networks to generate realistic faces from famous paintings like the Mona Lisa and American Gothic.
He uses neural networks to make the recreations as realistic as he can. If you want a quick break from work, check out all of Denis’ photorealistic art.
What are Neural Networks?
The Cambridge Dictionary describes a neural network as a computer system or a type of computer program that is designed to copy the way in which the human brain operates.
Ok, so I lied. I do want to say a couple of things about neural networks in demand forecasting…
Neural networks are highly useful in demand forecasting due to their ability to accommodate non-linear data and to capture subtle functional relationships among empirical data, even where the underlying relationships are unknown or hard to describe. Pretty good description of a demand planner’s every day, wouldn’t you say?
At Blue Ridge, we use neural network algorithms to dramatically improve forecast accuracy across hundreds of thousands of products. Neural networks help build, refine and intelligently optimize the forecast by:
- Harmonizing strategy across forecasting and inventory optimization
- Auto-aligning optimized parameters and settings to achieve business goals
- Providing one stop-point to know the state of business, deviation from goals, root cause identification and resolution recommendations
- Acting as a digital buddy to the buyer/planner that constantly learns and adapts to the business needs
- Allowing for configurable levels of automation, leveraging past user actions, tracking goals and delivering actionable intelligence