Distributed Learning vs Centralized Learning
The two main approaches to train models based on varying factors such as computational workload and modularity.
5 min read
Ramachandra Bharadwaj
Research Intern
Traditional Drift Detection
A brief summary and comparison of various drift detection algorithms like KSWIN, ADWIN and Page-Hinkley Test
4 min read
Ramachandra Bharadwaj
Research Intern
Drift Detection in Federated Learning Environment
Enhanced and Dynamic algorithms to accurately detect Drift in a distributed environment amongst clients where the traditional methods fail.
6 min read
Ramachandra Bharadwaj
Research Intern
How Federated learning is revolutionizing User's privacy
Federated Learning enhances user privacy and security, but it faces significant challenges related to data handling, system operations, communication, privacy, security, model aggregation, scalability, algorithmic tuning, regulatory compliance, and evaluation
4 min read
Akshay Nagamalla
Research Intern
Deep Dive into TFLite on-device training
TensorFlow Lite empowers on-device machine learning by enabling models to run directly on user devices, enhancing performance, privacy, and personalization.
8 min read
Akshay Nagamalla
Research Intern
Challenges faced in Federated Learning
Federated Learning, a cutting-edge approach in machine learning, offers transformative benefits but also faces significant challenges ranging from data heterogeneity to regulatory compliance.
5 min read
Akshay Nagamalla
Research Intern