Accessible Machine Learning

reportActive / Technical Report | Accesssion Number: AD1228492 | Open PDF

Abstract:

The project aims to develop an automated, interactive model discovery system that will enable users with no data science background to create and maintain useful models of complex processes, and that will increase effectiveness of competent data scientists via automation of multiple component tasks.

Author(s):
Dubrawski, Artur
Author Organization(s):
CARNEGIE-MELLON UNIV PITTSBURGH PA
Funding Organization(s):
AIR FORCE RESEARCH LAB ROME NY, ROME, NY
Document Type:
Technical Report/Final Report
Publication Date:
2024 May 23
Pagination:
21

Security Markings

DOCUMENT & CONTEXTUAL SUMMARY

Distribution Code:
A - Approved For Public Release
Distribution Statement: Public Release.
Copyright: Not Copyrighted

RECORD

Collection: TRECMS
Identifying Numbers
Contract Number(s):
FA8750-17-2-0130
 Subject Terms
Keyword(s):
interactive model discovery frameworks, ml(Machine learning), ml on complex objects, Automated composition of complex models, Dialog-capable Data Science Agents, Hyper-vised learning, Hyperactive Learning, Explainable Machine Learning, human interpretability of data models and results, scalable bayesian optimization, D3M(Data-Driven Discovery of Models)
Subject Categories:
Mathematical and Computer Sciences
Creation Date:
2024 May 23
Update Date:
2024 May 24