Keywords
Software Engineering, Artificial Intelligence, Software Architecture, Machine Learning, Self-adaptation, Deep learning, Reinforcement Learning, self-learning architectures, SA4ML, ML4SA
Objective
The objective of the research area is two-fold: i) Use a combination of machine learning techniques to enable software systems to overcome any runtime uncertainties (such as dynamic resource demands/resource constraints, failures, performance bottlenecks, etc.) through self-adaptation mechanisms.; ii) Develop software engineering/architecting practices for better development of ML-enabled systems by addressing key concerns like data quality, model versioning, team collaboration, self-adaptation, etc
Description
Modern software systems generates tremendous amount of data but face different architectural challenges arising from various concerns such as reliability, security, availability, resource constraints, etc. Some of those challenges can be solved using AI and on the other hand, we have AI systems that thrive on data but require better architecting practices to handle challenges associated to data quality, model versioning, model deployment, etc.. This combination of challenges in the field of Software architecture and AI has resulted in two broad research areas: i) Software architecture for AI systems. It primarily focuses on developing architectural techniques for better development of AI systems; ii) AI for Software architectures, which focuses on developing AI techniques to better architect software systems. This line of research focuses on addressing these two broad research challenges. This is accomplished by: i) by identifying the challenges and solutions for coming up with best practices for architecting AI in particular ML-enabled systems and; ii) By using a combination of deep learning and reinforcement learning techniques combined with quantitative verification mechanisms to enable software to continuously adapt and improve their architectures in runtime. The research is further applied to domains such as Microservice based systems, CPS (IoT, Robotics, etc). Detailed description of three focus areas are provided below.
Focus Areas
Selected Publications
- SWITCH: An Exemplar for Evaluating Self-Adaptive ML-Enabled Systems.on December 31, 2023 at 11:00 pm
Arya Marda, Shubham Kulkarni, Karthik Vaidhyanathan:SWITCH: An Exemplar for Evaluating Self-Adaptive ML-Enabled Systems. SEAMS@ICSE 2024: 143-149
- EcoMLS: A Self-Adaptation Approach for Architecting Green ML-Enabled Systems.on December 31, 2023 at 11:00 pm
Meghana Tedla, Shubham Kulkarni, Karthik Vaidhyanathan:EcoMLS: A Self-Adaptation Approach for Architecting Green ML-Enabled Systems. ICSA-C 2024: 230-237
- Architecting Digital Twin for Smart City Systems: A Case Study.on December 31, 2023 at 11:00 pm
Likhith Kanigolla, Gaurav Pal, Karthik Vaidhyanathan, Deepak Gangadharan, Anuradha Vattem:Architecting Digital Twin for Smart City Systems: A Case Study. ICSA-C 2024: 326-334
- Towards Architecting Sustainable MLOps: A Self-Adaptation Approach.on December 31, 2023 at 11:00 pm
Hiya Bhatt, Shrikara Arun, Adyansh Kakran, Karthik Vaidhyanathan:Towards Architecting Sustainable MLOps: A Self-Adaptation Approach. ICSA-C 2024: 179-182
- Reimagining Self-Adaptation in the Age of Large Language Models.on December 31, 2023 at 11:00 pm
Raghav Donakanti, Prakhar Jain, Shubham Kulkarni, Karthik Vaidhyanathan:Reimagining Self-Adaptation in the Age of Large Language Models. ICSA-C 2024: 171-174
- Leveraging Generative AI for Architecture Knowledge Management.on December 31, 2023 at 11:00 pm
Rudra Dhar, Karthik Vaidhyanathan, Vasudeva Varma:Leveraging Generative AI for Architecture Knowledge Management. ICSA-C 2024: 163-166
- SWITCH: An Exemplar for Evaluating Self-Adaptive ML-Enabled Systems.on December 31, 2023 at 11:00 pm
Arya Marda, Shubham Kulkarni, Karthik Vaidhyanathan:SWITCH: An Exemplar for Evaluating Self-Adaptive ML-Enabled Systems. CoRR abs/2402.06351 (2024)
- Can LLMs Generate Architectural Design Decisions? -An Exploratory Empirical study.on December 31, 2023 at 11:00 pm
Rudra Dhar, Karthik Vaidhyanathan, Vasudeva Varma:Can LLMs Generate Architectural Design Decisions? -An Exploratory Empirical study. CoRR abs/2403.01709 (2024)
- Exploratory Study of oneM2M-Based Interoperability Architectures for IoT: A Smart City Perspective.on December 31, 2023 at 11:00 pm
VJS Pranavasri, Leo Francis, Gaurav Pal, Ushasri Mogadali, Anuradha Vattem, Karthik Vaidhyanathan, Deepak Gangadharan:Exploratory Study of oneM2M-Based Interoperability Architectures for IoT: A Smart City Perspective. ICSA-C 2024: 16-23
- EcoMLS: A Self-Adaptation Approach for Architecting Green ML-Enabled Systems.on December 31, 2023 at 11:00 pm
Meghana Tedla, Shubham Kulkarni, Karthik Vaidhyanathan:EcoMLS: A Self-Adaptation Approach for Architecting Green ML-Enabled Systems. CoRR abs/2404.11411 (2024)