The 6th workshop on Neural Scaling Laws

This is the 6th workshop in our Scaling workshop series that started in Oct 2021. The objective of these workshop series, organized by the CERC in Autonomous AI Lab led by Irina Rish at the Universite de Montreal and Mila - Quebec AI Institute, is to provide a forum for discussing recent advances in foundation models - large-scale neural networks, pretrained in an unsupervised way on large and diverse datasets. Rapid increase in generalization capabilities of such models is an important step towards a long-standing objective of achieving Artificial General Intelligence (AGI), namely, a truly broad, versatile AI as opposed to "narrow" specialist.  Pushing the capabilities of the state-of-the-art AI systems towards AGI, while aiming to better understand and steer AI's behaviors towards those aligned with human intentions is the high-level objective of the  CERC in Autonomous AI program that is the initiator and primary driver behind this workshop series.  The topics of this workshop include (but are not limited to):

• empirical scaling laws of neural nets  behaviors, including  both capabilities and alignment metrics,  with increasing compute, model size, and pretraining data;

• capabilities vs alignment trade-off ? or capabilities+alignment  synergy?  

• advances in open-source AI, current trends and implications for the future