In this paper, we investigate the ability of large language models (LLMs) to suggest functionally correct, performance improving code edits.
In this paper, we study the OOD generalization of neural algorithmic reasoning tasks, where the goal is to learn an algorithm (e. g., sorting, breadth-first search, and depth-first search) from input-output pairs using deep neural networks.
Neural fields have rapidly been adopted for representing 3D signals, but their application to more classical 2D image-processing has been relatively limited.
To automatically learn these hints from the dataset, we propose a novel discrete variational auto-encoder, where each discrete latent variable represents a different learned category of code-edit that increases performance.
Machine Learning has been successfully applied in systems applications such as memory prefetching and caching, where learned models have been shown to outperform heuristics.
An alternative paradigm is to use a "data-driven", offline approach that utilizes logged simulation data, to architect hardware accelerators, without needing any form of simulations.
We are the first to take a unified perspective to jointly explain the oversmoothing and heterophily problems at the node level.
We propose a general and scalable approximate sampling strategy for probabilistic models with discrete variables.
no code implementations • 2 Feb 2021 • Amir Yazdanbakhsh, Christof Angermueller, Berkin Akin, Yanqi Zhou, Albin Jones, Milad Hashemi, Kevin Swersky, Satrajit Chatterjee, Ravi Narayanaswami, James Laudon
We further show that by transferring knowledge between target architectures with different design constraints, Apollo is able to find optimal configurations faster and often with better objective value (up to 25% improvements).
Energy-Based Models (EBMs) present a flexible and appealing way to represent uncertainty.
The use of deep learning has grown at an exponential rate, giving rise to numerous specialized hardware and software systems for deep learning.
While directly applying Belady's is infeasible since the future is unknown, we train a policy conditioned only on past accesses that accurately approximates Belady's even on diverse and complex access patterns, and call this approach Parrot.
A significant effort has been made to train neural networks that replicate algorithmic reasoning, but they often fail to learn the abstract concepts underlying these algorithms.
In this work, we propose a new approach to use GNNs to learn fused representations of general source code and its execution.
In this paper, we demonstrate the potential of deep learning to address the von Neumann bottleneck of memory performance.