Automated Configuration Synthesis for Machine Learning Models: A git-Based Requirement and Architecture Management System
- URL: http://arxiv.org/abs/2404.17244v1
- Date: Fri, 26 Apr 2024 08:35:02 GMT
- Title: Automated Configuration Synthesis for Machine Learning Models: A git-Based Requirement and Architecture Management System
- Authors: Abdullatif AlShriaf, Hans-Martin Heyn, Eric Knauss,
- Abstract summary: This work introduces a tool for generating runtime configurations automatically from textual requirements stored as artifacts in git repositories (a.k.a. T-Reqs) alongside the software code.
The tool leverages T-Reqs-modelled architectural description to identify relevant configuration properties for the deployment of artificial intelligence (AI)-enabled software systems.
- Score: 5.095988654970361
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: This work introduces a tool for generating runtime configurations automatically from textual requirements stored as artifacts in git repositories (a.k.a. T-Reqs) alongside the software code. The tool leverages T-Reqs-modelled architectural description to identify relevant configuration properties for the deployment of artificial intelligence (AI)-enabled software systems. This enables traceable configuration generation, taking into account both functional and non-functional requirements. The resulting configuration specification also includes the dynamic properties that need to be adjusted and the rationale behind their adjustment. We show that this intermediary format can be directly used by the system or adapted for specific targets, for example in order to achieve runtime optimisations in term of ML model size before deployment.
Related papers
- Control Industrial Automation System with Large Language Models [2.2369578015657954]
This paper introduces a framework for integrating large language models with industrial automation systems.
At the core of the framework are an agent system designed for industrial tasks, a structured prompting method, and an event-driven information modeling mechanism.
Our contribution includes a formal system design, proof-of-concept implementation, and a method for generating task-specific datasets.
arXiv Detail & Related papers (2024-09-26T16:19:37Z) - Towards Living Software Architecture Diagrams [0.0]
We propose a tool that generates architectural diagrams for a software system by analyzing its software artifacts and unifying them into a comprehensive system representation.
This representation can be manually modified while ensuring that changes are reintegrated into the diagram when it is regenerated.
arXiv Detail & Related papers (2024-07-25T12:31:52Z) - Towards Single-System Illusion in Software-Defined Vehicles -- Automated, AI-Powered Workflow [3.2821049498759094]
We propose a novel model- and feature-based approach to development of vehicle software systems.
One of the key points of the presented approach is the inclusion of modern generative AI, specifically Large Language Models (LLMs)
The resulting pipeline is automated to a large extent, with feedback being generated at each step.
arXiv Detail & Related papers (2024-03-21T15:07:57Z) - Exploiting Contextual Target Attributes for Target Sentiment
Classification [53.30511968323911]
Existing PTLM-based models for TSC can be categorized into two groups: 1) fine-tuning-based models that adopt PTLM as the context encoder; 2) prompting-based models that transfer the classification task to the text/word generation task.
We present a new perspective of leveraging PTLM for TSC: simultaneously leveraging the merits of both language modeling and explicit target-context interactions via contextual target attributes.
arXiv Detail & Related papers (2023-12-21T11:45:28Z) - Ensemble-based modeling abstractions for modern self-optimizing systems [0.0]
We extend our ensemble-based component model DEECo with the capability to use machine-learning and Industry optimizations in establishing and reconfiguration of component ensembles.
We argue that machine-learning and optimizations is a key feature for modern smart systems which are to learn over the time and optimize their behavior at runtime to deal with uncertainty in their environment.
arXiv Detail & Related papers (2023-09-11T21:01:11Z) - PDSketch: Integrated Planning Domain Programming and Learning [86.07442931141637]
We present a new domain definition language, named PDSketch.
It allows users to flexibly define high-level structures in the transition models.
Details of the transition model will be filled in by trainable neural networks.
arXiv Detail & Related papers (2023-03-09T18:54:12Z) - ROAD-R: The Autonomous Driving Dataset with Logical Requirements [54.608762221119406]
We introduce the ROad event Awareness dataset with logical Requirements (ROAD-R)
ROAD-R is the first publicly available dataset for autonomous driving with requirements expressed as logical constraints.
We show that it is possible to exploit them to create models that (i) have a better performance, and (ii) are guaranteed to be compliant with the requirements themselves.
arXiv Detail & Related papers (2022-10-04T13:22:19Z) - Policy Architectures for Compositional Generalization in Control [71.61675703776628]
We introduce a framework for modeling entity-based compositional structure in tasks.
Our policies are flexible and can be trained end-to-end without requiring any action primitives.
arXiv Detail & Related papers (2022-03-10T06:44:24Z) - Federated Learning with Heterogeneous Architectures using Graph
HyperNetworks [154.60662664160333]
We propose a new FL framework that accommodates heterogeneous client architecture by adopting a graph hypernetwork for parameter sharing.
Unlike existing solutions, our framework does not limit the clients to share the same architecture type, makes no use of external data and does not require clients to disclose their model architecture.
arXiv Detail & Related papers (2022-01-20T21:36:25Z) - SOLIS -- The MLOps journey from data acquisition to actionable insights [62.997667081978825]
In this paper we present a unified deployment pipeline and freedom-to-operate approach that supports all requirements while using basic cross-platform tensor framework and script language engines.
This approach however does not supply the needed procedures and pipelines for the actual deployment of machine learning capabilities in real production grade systems.
arXiv Detail & Related papers (2021-12-22T14:45:37Z) - Amazon SageMaker Automatic Model Tuning: Scalable Black-box Optimization [23.52446054521187]
Amazon SageMaker Automatic Model Tuning (AMT) is a fully managed system for black-box optimization at scale.
AMT finds the best version of a machine learning model by repeatedly training it with different hyperparameter configurations.
It can be used with built-in algorithms, custom algorithms, and Amazon SageMaker pre-built containers for machine learning frameworks.
arXiv Detail & Related papers (2020-12-15T18:34:34Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.