#036 - Max Welling: Quantum, Manifolds & Symmetries in ML

Research Norms

Max Welling explores the need for a paradigm shift in research acceptance and review processes. He advocates for an open review system where reviews hold as much weight as the papers themselves, fostering a more dynamic and continuous research environment. This approach aims to reduce the demotivation faced by students when their non-mainstream ideas are repeatedly rejected by traditional conferences 1.

It's much more like a marketplace where ideas go around, conferences come in and ask you to publish things, and it's just you then present it, and then you can just continue with your research or stop it and go to a new piece of work or something like this.

--- Max Welling

Welling believes that such a system would encourage originality and provide a platform for diverse ideas to flourish 2.

   

Innovative Models

Max Welling introduces groundbreaking models in machine learning, including probabilistic numeric convolutional neural networks and quantum deformed neural networks. These models address challenges like irregularly sampled data and leverage quantum mechanics to enhance machine learning capabilities 3.

You can think of quantum mechanics as another theory of statistics in some sense.

--- Max Welling

Welling's work demonstrates how quantum principles can be applied to neural networks, potentially revolutionizing classical predictions and offering new insights into data processing 4.