Introducing CHITA, an optimization-based approach for pruning pre-trained neural networks at scale. Learn how it leverages advances from several fields and outperforms state-of-the-art pruning methods in terms of scalability and performance tradeoffs → https://t.co/Efokrj0y6W