Apache TVM is an open-source machine learning stack for CPUs, GPUs, and specialized accelerators. It aims to close the gap between the productivity-focused deep learning frameworks, and the performance- or efficiency-oriented hardware backends.

We're excited to announce the 2023 TVM Virtual Conference, March 13th - 17th. The TVM Conference will cover the state of the art of machine learning compilation and optimization. We welcome TVM contributors, new and current users, deep learning researchers, and practitioners from the broader community. The conference will discuss recent advances in frameworks, machine learning compilation, systems and architecture support, security, acceleration, and deployment.

This years schedule will be expanded to focus on tutorial sessions on March 13th to help introduce new users and developers to the project and community. Overall, we are seeking:

• Technical and research talks: novel uses and advances in machine learning compilation
• Business case studies: describe your firsthand experience how TVM is used in industry
• Lightning Talks: a quick overview of something relevant to TVM, with minimal technical depth
• Tutorials: a step-by-step walkthrough showing how to accomplish something that relies mostly on features committed to TVM's main branch

This form will take 5-10 minutes to complete.