2 citations · 4 across the 4 of their papers we have counts for
5 papers
TorchTPU - A Hybrid Eager/Compiled Architecture for Native PyTorch Execution on TPU
Claudio Basile, Michael Voznesensky, Gunhyun Park +50
TorchTPU is an open-source collaboration between Google and Meta that introduces a native and principled PyTorch backend for Google TPUs. It enables seamless out-of-the-box model e…
Guided Tensor Lifting
Yixuan Li, José Wesley de Souza Magalhães, Alexander Brauckmann +2
Domain-specific languages (DSLs) for machine learning are revolutionizing the speed and efficiency of machine learning workloads as they enable users easy access to high-performanc…
mlirSynth: Automatic, Retargetable Program Raising in Multi-Level IR using Program Synthesis
Alexander Brauckmann, Elizabeth Polgreen, Tobias Grosser +1
MLIR is an emerging compiler infrastructure for modern hardware, but existing programs cannot take advantage of MLIR's high-performance compilation if they are described in lower-l…
Rewriting History: Repurposing Domain-Specific CGRAs
Jackson Woodruff, Thomas Koehler, Alexander Brauckmann +3
Coarse-grained reconfigurable arrays (CGRAs) are domain-specific devices promising both the flexibility of FPGAs and the performance of ASICs. However, with restricted domains come…
A Reinforcement Learning Environment for Polyhedral Optimizations
Alexander Brauckmann, Andrés Goens, Jeronimo Castrillon
The polyhedral model allows a structured way of defining semantics-preserving transformations to improve the performance of a large class of loops. Finding profitable points in thi…