First Conference Paper Accepted

🎉 Huge milestone alert! My first conference paper has been accepted to NeurIPS 2024 (Conference on Neural Information Processing Systems)!

Paper Details

Title: “Federated Continual Learning with Privacy-Preserving Knowledge Distillation”

Abstract: We propose a novel framework that enables continual learning in federated settings while preserving privacy and preventing catastrophic forgetting.

Key Contributions

1. Novel Architecture: Developed a privacy-preserving knowledge distillation mechanism
2. Theoretical Analysis: Provided convergence guarantees for federated continual learning
3. Empirical Validation: Demonstrated superior performance on benchmark datasets
4. Real-world Application: Showed practical benefits in healthcare and finance domains

Research Journey

This paper represents months of:

• Industry Problem Identification: Recognizing privacy challenges in financial ML
• Academic Collaboration: Working with Georgia Tech researchers during evenings/weekends
• Algorithm Development: Designing solutions that work in real trading environments
• Rigorous Validation: Testing on both academic benchmarks and financial datasets

Impact and Recognition

• Acceptance Rate: NeurIPS 2024 had an acceptance rate of ~25%, making this particularly meaningful
• Industry-Academic Bridge: Joint work combining industry insights with academic rigor
• Practical Applications: Framework already being evaluated for production deployment at IB
• Future Collaboration: Opening doors for more industry-academic research partnerships

Balancing Industry and Research

This achievement demonstrates that:

• Full-time Industry Work doesn’t preclude meaningful research contributions
• Real-world Problems can drive valuable academic research
• Industry Experience provides unique perspectives for research innovation
• Evening/Weekend Research can yield high-impact results

Gratitude

Huge thanks to Prof. Zsolt Kira and the Georgia Tech research team for welcoming industry collaboration, my colleagues at Interactive Brokers for supporting research interests, and the amazing ML community for fostering industry-academic partnerships!

Looking forward to presenting at the conference and showing how industry insights can advance academic research! 🚀