Algorithms Data Science Generative Adversarial Networks Machine-Learning
For hundreds of years, scientists were developing strong theories and rigorous mathematical models to explain patterns and dependencies in data and processes around us. Today instead of modeling detailed features of the data by ourselves we rely on deep neural networks and they don't let us down. So, the natural question arises: do we really need human experts to describe the world mathematically or let's just let AI do all the work?
In this session, we will connect dots between generative neural architectures (GANs, VAEs, and others) and mathematical models like ODEs (ordinary differential equations) for different applications. We analyze empirically what human experts did and what neural networks have learned by themselves and will try to understand, how close we are to fully rely on AI as we rely on science in the most important areas: medicine, military, finance, space tech, and many others.
This session planned to interactive: we will review theory first, and after will execute the code with generative neural networks on different datasets to check the hypothesis of the "neural mathematician" by ourselves. Some previous experience or knowledge of neural networks is preferred. Coding part will include snippets with PyTorch, but knowledge of this framework is nor required.
Type: Interactive (60 mins); Python level: Advanced; Domain level: Beginner
I am Alex Honchar, machine learning expert with more than 5 years of varied experience in the field. At the moment I am a co-founder and ML architect in Mawi Solutions, where I am working on biosignal analysis and co-founder and CTO of Neurons Lab, where we're enhancing businesses with new data-driven features and products.
Meanwhile, I am also writing in a popular blog on Medium and giving public speeches on conferences and meetups in Ukraine, Italy, and Spain. My mission is to build AI products, that truly outperform human skills or deliver new knowledge that couldn’t be discovered solely with human intellect. Of course, I believe that such products should not leave professionals without their jobs, but the opposite - become valuable partners for achieving more ambitious goals.