The rapid growth in the size of deep learning models strains the capabilities of traditional dense computation paradigms. Leveraging sparse computation has become increasingly popular for training and deploying large-scale models, but existing deep learning frameworks lack extensive support for sparse operations. To bridge this gap, we introduce Scorch, a library that seamlessly integrates efficient sparse tensor computation into the PyTorch ecosystem, with an initial focus on inference workloads on CPUs. Scorch provides a flexible and intuitive interface for sparse tensors, supporting diverse sparse data structures. Scorch introduces a compiler stack that automates key optimizations, including automatic loop ordering, tiling, and format inference. Combined with a runtime that adapts its execution to both dense and sparse data, Scorch delivers substantial speedups over hand-written PyTorch Sparse (torch.sparse) operations without sacrificing usability. More importantly, Scorch enables efficient computation of complex sparse operations that lack hand-optimized PyTorch implementations. This flexibility is crucial for exploring novel sparse architectures. We demonstrate Scorch's ease of use and performance gains on diverse deep learning models across multiple domains. With only minimal code changes, Scorch achieves 1.05-5.78x speedups over PyTorch Sparse on end-to-end tasks. Scorch's seamless integration and performance gains make it a valuable addition to the PyTorch ecosystem. We believe Scorch will enable wider exploration of sparsity as a tool for scaling deep learning and inform the development of other sparse libraries.

When Jason West and Vince Zampella set up Respawn Entertainment in 2010, they had one ambition: to produce a new first-person shooter that would have as massive an impact on the genre as their previous creation: the Call of Duty series. It was a big ask, but when Titanfall arrived three years later, the game was certainly a brilliant attempt. The sci-fi shooter boasted an innovative mechanic allowing players to summon a giant robot into the arena, and an incredibly fluid, free-running movement style – all combined into a set of blisteringly loud and detailed map designs. But one thing many players said about Titanfall was that, beyond the raw speed and inarguable thrill of the highly vertical, highly acrobatic gameplay, there was little in the way of tactical depth. It's something the team says it wants to address. "We learned a lot from the community," says producer Drew McCoy.