Related papers: No Saved Kaleidosope: an 100% Jitted Neural Networ…
This paper presents an open-source neural machine translation toolkit named CytonMT (https://github.com/arthurxlw/cytonMt). The toolkit is built from scratch only using C++ and NVIDIA's GPU-accelerated libraries. The toolkit features…
Python is a particularly appealing language to carry out data analysis, owing in part to its user-friendly character as well as its access to well maintained and powerful libraries like NumPy and SciPy. Still, for the purpose of analyzing…
Training modern deep learning models is increasingly constrained by GPU memory and compute limits. While Randomized Numerical Linear Algebra (RandNLA) offers proven techniques to compress these models, the lack of a unified,…
Path signatures provide a rich representation of sequential data, with strong theoretical guarantees and good performance in a variety of machine-learning tasks. While signatures have progressed from fixed feature extractors to trainable…
Modern neural network architectures use structured linear transformations, such as low-rank matrices, sparse matrices, permutations, and the Fourier transform, to improve inference speed and reduce memory usage compared to general linear…
Unconstrained handwritten text recognition is a major step in most document analysis tasks. This is generally processed by deep recurrent neural networks and more specifically with the use of Long Short-Term Memory cells. The main drawbacks…
Modern machine learning frameworks are complex: they are typically organised in multiple layers each of which is written in a different language and they depend on a number of external libraries, but at their core they mainly consist of…
The prohibitive expense of automatic performance tuning at scale has largely limited the use of autotuning to libraries for shared-memory and GPU architectures. We introduce a framework for approximate autotuning that achieves a desired…
Hardware failures are a growing challenge for machine learning accelerators, many of which are based on systolic arrays. When a permanent hardware failure occurs in a systolic array, existing solutions include localizing and isolating the…
This paper proposes an adaptive neural-compilation framework to address the problem of efficient program learning. Traditional code optimisation strategies used in compilers are based on applying pre-specified set of transformations that…
We present a technique for applying (forward and) reverse-mode automatic differentiation (AD) on a non-recursive second-order functional array language that supports nested parallelism and is primarily aimed at efficient GPU execution. The…
Existing memory systems for language agents address memory management: how to retrieve and page more information within a context budget. We address a complementary problem -- memory utility: what experience is worth keeping, and how it…
Language implementation frameworks, e.g., RPython and Truffle/Graal, are practical tools for creating efficient virtual machines, including a well-functioning just-in-time (JIT) compiler. It is demanding to support multitier JIT compilation…
Neural-networks-driven intelligent data-plane (NN-driven IDP) is becoming an emerging topic for excellent accuracy and high performance. Meanwhile we argue that NN-driven IDP should satisfy three design goals: the flexibility to support…
Convolutional neural networks (CNNs) are the core of most state-of-the-art deep learning algorithms specialized for object detection and classification. CNNs are both computationally complex and embarrassingly parallel. Two properties that…
As LiDAR sensors have become ubiquitous, the need for an efficient LiDAR data compression algorithm has increased. Modern LiDARs produce gigabytes of scan data per hour and are often used in applications with limited compute, bandwidth, and…
As a language model that integrates traditional symbolic operations and flexible neural representations, recurrent neural network grammars (RNNGs) have attracted great attention from both scientific and engineering perspectives. However,…
Scientific machine learning often requires combining known physics with unknown parameters or correction terms learned from data. Existing approaches either ignore known structure, encode it as a soft penalty, or require hand-written…
Large language models (LLMs) excel in program synthesis, yet their capacity for neural architecture design -- balancing syntactic reliability, performance, and structural novelty -- remains underexplored. We present a closed-loop…
A major challenge in the deployment of scientific software solutions is the adaptation of research prototypes to production-grade code. While high-level languages like MATLAB are useful for rapid prototyping, they lack the resource…