REVIEW 8 cited by
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Tensor Comprehensions: Framework-Agnostic High-Performance Machine Learning Abstractions
read the original abstract
Deep learning models with convolutional and recurrent networks are now ubiquitous and analyze massive amounts of audio, image, video, text and graph data, with applications in automatic translation, speech-to-text, scene understanding, ranking user preferences, ad placement, etc. Competing frameworks for building these networks such as TensorFlow, Chainer, CNTK, Torch/PyTorch, Caffe1/2, MXNet and Theano, explore different tradeoffs between usability and expressiveness, research or production orientation and supported hardware. They operate on a DAG of computational operators, wrapping high-performance libraries such as CUDNN (for NVIDIA GPUs) or NNPACK (for various CPUs), and automate memory allocation, synchronization, distribution. Custom operators are needed where the computation does not fit existing high-performance library calls, usually at a high engineering cost. This is frequently required when new operators are invented by researchers: such operators suffer a severe performance penalty, which limits the pace of innovation. Furthermore, even if there is an existing runtime call these frameworks can use, it often doesn't offer optimal performance for a user's particular network architecture and dataset, missing optimizations between operators as well as optimizations that can be done knowing the size and shape of data. Our contributions include (1) a language close to the mathematics of deep learning called Tensor Comprehensions, (2) a polyhedral Just-In-Time compiler to convert a mathematical description of a deep learning DAG into a CUDA kernel with delegated memory management and synchronization, also providing optimizations such as operator fusion and specialization for specific sizes, (3) a compilation cache populated by an autotuner. [Abstract cutoff]
Forward citations
Cited by 8 Pith papers
-
Toward Compiler World Models: Learning Latent Dynamics for Efficient Tensor Program Search
A latent dynamics model for schedule trajectories in TVM AutoScheduler finds programs with 1.37x better GPU latency than Ansor using the same 64 trials and matches 10K-trial Ansor with 10x fewer measurements.
-
ScanWeaver: Compiler-Driven Parallelization of Affine Recurrences via Associative Scan Lowering
ScanWeaver lowers affine recurrences to compiler-generated Blelloch scans in MLIR, producing executable GPU code validated on selective-scan workloads against PyTorch, CUDA, and Mamba baselines.
-
Step-TP: A Grounded, Step-Level Dataset with Chain-of-Thought Reasoning for LLM-Guided Tensor Program Optimization
Step-TP is a dataset providing grounded, atomic step-level IR transitions and CoT supervision to enable reliable multi-step LLM-guided tensor program optimization instead of end-to-end imitation.
-
Mat2Boundary: Treating User-Defined Boundary Condition as SpMV for Distributed PDE Solvers on Block-Structured Grids
Mat2Boundary treats boundary conditions as sparse matrix-vector products and uses multi-stage compilation with polyhedral analysis to generate efficient matrix-free kernels and communication schedules for distributed ...
-
Unraveling the Key of Machine Learning-based Android Malware Detection
A taxonomy and re-implementation study of 12 ML Android malware detectors finds persistent vulnerabilities to malware evolution and adversarial attacks due to insufficient capture of malware semantics.
-
Neural Computers
Neural Computers are introduced as a new machine form where computation, memory, and I/O are unified in a learned runtime state, with initial video-model experiments showing acquisition of basic interface primitives f...
-
Space Filling Curves is All You Need: Communication-Avoiding Matrix Multiplication Made Simple
Space-filling curves enable platform- and shape-oblivious communication-avoiding matrix multiplication that outperforms vendor libraries by up to 5.5x on CPUs while also accelerating LLM prefill and distributed workloads.
-
DMuon: Efficient Distributed Muon Training with Near-Adam Overhead
DMuon delivers 1.48x-3.01x end-to-end and 6.85x-163x optimizer-step speedups for Muon on embodied foundation models and LLMs while matching AdamW per-step latency.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.