Publications
Research on Large-Scale Adaptive Distributed Systems
Under SubmissionTowards Full Stack Adaptivity in Permissioned Blockchains
VLDB '24
A comprehensive approach to making all layers of the blockchain stack adaptive to changing conditions. This research demonstrates how machine learning can be applied across different blockchain components to optimize performance under varying workloads.
Towards Adaptive Fault-Tolerant Sharded Databases
AIDB @ VLDB '23
Novel techniques for making sharded database systems resilient to failures while maintaining adaptivity. This paper presents algorithms that allow database systems to adjust their behavior in response to node failures without sacrificing performance.
AdaChain: A Learned Adaptive Blockchain
VLDB '23
A blockchain system that uses machine learning to adapt to changing workloads and network conditions. AdaChain represents a breakthrough in blockchain adaptivity by automatically tuning system parameters to maintain optimal performance as conditions change.
High Speed SRT Divider for Intelligent Embedded System
arXiv preprint (2018)
This paper presents an optimized implementation of the SRT division algorithm for embedded systems, improving computational efficiency for resource-constrained devices.
Human action recognition system using good features and multilayer perceptron network
International Conference on Communication and Signal Processing (ICCSP) 2017
A novel approach to human action recognition using optimized feature extraction techniques and neural networks, achieving high accuracy with computational efficiency.