arXiv subjects

Andrew Adams

Publications and source records attributed to Andrew Adams.

7 recordsLinked to original sources

arXiv2026

Finding Fast Filters

Processing images, video, and audio often requires running large finite impulse response (FIR) filters with strict performance and latency requirements. Prior methods for fast filter approximations are special cases or combinations of a few key techniques: multi-rate and recurrent filtering, and decomposing filters into sums or cascades. We unify these techniques as primitives within a single design language for fast 1D and 2D filters. Given a target filter to approximate, we automatically search this program space, fitting continuous parameters with gradient descent, to generate a Pareto frontier of algorithms that trade off performance with quality. Our system produces substantially higher-quality and faster filter approximations than have been previously described for several popular imaging and audio filters. Furthermore we demonstrate how to automatically lower programs in this design space to optimized, vectorized, parallel, C++ code which is fused for data locality.

cs.GR↗

arXiv2025

Pushing Tensor Accelerators Beyond MatMul in a User-Schedulable Language

Tensor accelerators now represent a growing share of compute resources in modern CPUs and GPUs. However, they are hard to program, leading developers to use vendor-provided kernel libraries that support tensor accelerators. As a result, the usage of tensor accelerators is limited to the provided interface, mainly designed for traditional ML and scientific computing workloads. In this paper, we show that tensor accelerators can improve the performance of applications beyond simple variants of MatMul. For example, many image processing pipelines are linear transformations over matrices in disguise and can therefore utilize such specialized hardware. This is nonetheless hindered by the difficulties in programming tensor accelerators. We tackle this problem with compiler-based techniques. We use the Halide user-schedulable language and express operations as Halide algorithms succinctly. To this end, we implement a flexible tensor instruction selector based on equality saturation. The tensor instruction selector supports both CPU- and GPU-attached tensor accelerators and works with existing scheduling operations (e.g., producer-consumer fusion). Together, this enables developers to write diverse accelerator-leveraging applications in a few dozen lines. Using our system, we demonstrate the potential of tensor accelerators beyond their traditional domains. We implement several image processing pipelines (e.g., filtering, resampling, and denoising) in our system and evaluate them against non-accelerator-leveraging baselines. We show that these pipelines can achieve significant speedups. For example, a downsampling routine is sped up by $6.1\times$ by utilizing Tensor Cores on an Nvidia RTX 4070 GPU.

cs.PL↗

arXiv2025

Bonsai: Compiling Queries to Pruned Tree Traversals

Trees can accelerate queries that search or aggregate values over large collections. They achieve this by storing metadata that enables quick pruning (or inclusion) of subtrees when predicates on that metadata can prove that none (or all) of the data in a subtree affect the query result. Existing systems implement this pruning logic manually for each query predicate and data structure. We generalize and mechanize this class of optimization. Our method derives conditions for when subtrees can be pruned (or included wholesale), expressed in terms of the metadata available at each node. We efficiently generate these conditions using symbolic interval analysis, extended with new rules to handle geometric predicates (e.g., intersection, containment). Additionally, our compiler fuses compound queries (e.g., reductions on filters) into a single tree traversal. These techniques enable the automatic derivation of generalized single-index and dual-index tree joins that support a wide class of join predicates beyond standard equality and range predicates. The generated traversals match the behavior of expert-written code that implements query-specific traversals, and can asymptotically outperform the linear scans and nested-loop joins that existing systems fall back to when hand-written cases do not apply.

cs.PL↗

arXiv2022

Persistent Weak Interferer Detection in WiFi Networks: A Deep Learning Based Approach

In this paper, we explore the use of multiple deep learning techniques to detect weak interference in WiFi networks. Given the low interference signal levels involved, this scenario tends to be difficult to detect. However, even signal-to-interference ratios exceeding 20 dB can cause significant throughput degradation and latency. Furthermore, the resultant packet error rate may not be enough to force the WiFi network to fallback to a more robust physical layer configuration. Deep learning applied directly to sampled radio frequency data has the potential to perform detection much cheaper than successive interference cancellation, which is important for real-time persistent network monitoring. The techniques explored in this work include maximum softmax probability, distance metric learning, variational autoencoder, and autoreggressive log-likelihood. We also introduce the notion of generalized outlier exposure for these techniques, and show its importance in detecting weak interference. Our results indicate that with outlier exposure, maximum softmax probability, distance metric learning, and autoreggresive log-likelihood are capable of reliably detecting interference more than 20 dB below the 802.11 specified minimum sensitivity levels. We believe this presents a unique software solution to real-time, persistent network monitoring.

eess.SP↗

arXiv2020

Efficient Automatic Scheduling of Imaging and Vision Pipelines for the GPU

We present a new algorithm to quickly generate high-performance GPU implementations of complex imaging and vision pipelines, directly from high-level Halide algorithm code. It is fully automatic, requiring no schedule templates or hand-optimized kernels. We address the scalability challenge of extending search-based automatic scheduling to map large real-world programs to the deep hierarchies of memory and parallelism on GPU architectures in reasonable compile time. We achieve this using (1) a two-phase search algorithm that first 'freezes' decisions for the lowest cost sections of a program, allowing relatively more time to be spent on the important stages, (2) a hierarchical sampling strategy that groups schedules based on their structural similarity, then samples representatives to be evaluated, allowing us to explore a large space with few samples, and (3) memoization of repeated partial schedules, amortizing their cost over all their occurrences. We guide the process with an efficient cost model combining machine learning, program analysis, and GPU architecture knowledge. We evaluate our method's performance on a diverse suite of real-world imaging and vision pipelines. Our scalability optimizations lead to average compile time speedups of 49x (up to 530x). We find schedules that are on average 1.7x faster than existing automatic solutions (up to 5x), and competitive with what the best human experts were able to achieve in an active effort to beat our automatic results.

cs.PL↗

arXiv2019

Trusted CI Experiences in Cybersecurity and Service to Open Science

This article describes experiences and lessons learned from the Trusted CI project, funded by the US National Science Foundation to serve the community as the NSF Cybersecurity Center of Excellence. Trusted CI is an effort to address cybersecurity for the open science community through a single organization that provides leadership, training, consulting, and knowledge to that community. The article describes the experiences and lessons learned of Trusted CI regarding both cybersecurity for open science and managing the process of providing centralized services to a broad and diverse community.

cs.CR↗

arXiv2018

Tensor Comprehensions: Framework-Agnostic High-Performance Machine Learning Abstractions

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]

cs.PL↗