English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Code optimization is a daunting task that requires a significant level ofexpertise from experienced programmers. This level of expertise is notsufficient when compared to the rapid development of new hardwarearchitectures. Towards advancing the whole code optimization process, recentapproaches rely on machine learning and artificial intelligence techniques.This paper introduces a new framework to decrease the complexity of codeoptimization. The proposed framework builds on large language models (LLMs) andreinforcement learning (RL) and enables LLMs to receive feedback from theirenvironment (i.e., unit tests) during the fine-tuning process. We compare ourframework with existing state-of-the-art models and show that it is moreefficient with respect to speed and computational usage, as a result of thedecrement in training steps and its applicability to models with fewerparameters. Additionally, our framework reduces the possibility of logical andsyntactical errors. Toward evaluating our approach, we run several experimentson the PIE dataset using a CodeT5 language model and RRHF, a new reinforcementlearning algorithm. We adopt a variety of evaluation metrics with regards tooptimization quality, and speedup. The evaluation results demonstrate that theproposed framework has similar results in comparison with existing models usingshorter training times and smaller pre-trained models. In particular, weaccomplish an increase of 5.6% and 2.2 over the baseline models concerning the%OP T and SP metrics.

Leveraging Reinforcement Learning and Large Language Models for Code Optimization