[2102.09798] Training Neural Networks is $\exists\mathbb R$-complete

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs. arXiv Operational Status

[2102.09798] Training Neural Networks is $\exists\mathbb R$-complete Skip to main content Search arXiv Press Enter to search · Advanced search --> Computer Science > Computational Complexity arXiv:2102.09798 (cs) [Submitted on 19 Feb 2021 ( v1 ), last revised 19 Nov 2021 (this version, v2)] Title: Training Neural Networks is $\exists\mathbb R$-complete Authors: Mikkel Abrahamsen , Linda Kleist , Tillmann Miltzow View a PDF of the paper titled Training Neural Networks is $\exists\mathbb R$-complete, by Mikkel Abrahamsen and 2 other authors View PDF Abstract: Given a neural network, train

related reading

• Aman's AI Journal • Primers • Ilya Sutskever's Top 30aman.ai
• A Recipe for Training Neural Networkskarpathy.github.io
• Neural networks and deep learningneuralnetworksanddeeplearning.com
• Neural network training makes beautiful fractals | Jascha’s blogsohl-dickstein.github.io
• Zoom In: An Introduction to Circuitsdistill.pub
• On neural scaling and the quanta hypothesisericjmichaud.com
• A Recipe for Training Neural Networkskarpathy.github.io
• Machine Learning, Kolmogorov Complexity, and Squishy Bunniestheorangeduck.com
• Jane Street Blog - Can you reverse engineer our neural network?blog.janestreet.com
• arxiv.org/pdf/1805.08522arxiv.org
• [2604.21691] There Will Be a Scientific Theory of Deep Learningarxiv.org
• Deep learning as program synthesis — LessWronglesswrong.com