ÃÛÌÒ´«Ã½

Summary

Dr. Khushi is an Associate Professor in AI and has over 2 decades of industry and academic experience. He completed his PhD in AI and Data Science from the University of Sydney, Australia in which he developed novel algorithms for understanding big genomic data. During his PhD and later in postdoc (2014-2017) at the Children’s Medical Research Institute, Australia, he developed many automated AI-based algorithms for expediting medical conditions and lately has been working on UKRI NEC grant for establishing bioinformatics tools to identify fish sensitivity.

Dr. Khushi has supervised 6 PhD to completion and more than 100 postgraduate dissertations in various domains of AI and data mining and is always willing to supervise very talented PhD candidates.

Dr. Khushi's prowess in FinTech AI is evident in his numerous accolades, significant research grants, and over 80 peer-reviewed publications in prestigious academic journals. His groundbreaking work has received prestigious best research awards from esteemed outlets like IEEE Transactions on Computational Social Systems and PeerJ. His research contributions place him among the prestigious Stanford/Elsevier list of the top 2% of scientists globally. 

Examples of some of his significant FinTech work are as follows:

• He has implemented innovative financial models to improve their decision-making. One such example is his work with an international bank where Dr. Khushi developed a cutting-edge portfolio management model using reinforcement learning that optimizes cumulative returns and the Sharpe Ratio [1]. This model has been instrumental in refining the bank's investment strategies and risk management practices, leading to significantly more informed and profitable decisions. 
• Dr. Khushi's research for a major international bank addressed its need for robust risk assessment tools. He created sophisticated models that simulate realistic financial scenarios, including market crashes, empowering fund managers with deeper insights into risk dynamics. This allows them to make more resilient investment decisions during turbulent market periods [2]. Dr. Khushi's impact transcends specific projects. He has made significant strides in developing algorithms for modelling non-stationary time-series data. This work has not only advanced academic understanding but also provided practical tools for financial analysts and economists to better predict and respond to market changes [3, 4]. 
• Dr. Khushi's impact on quantitative finance extends beyond existing metrics. He identified limitations in popular risk-adjusted return measures like the Sharpe Ratio (sensitive to high volatility) and Sterling Ratio (punishes large drawdowns). To address these shortcomings, he invented the SS Ratio, incorporating both volatility and drawdown sensitivities. His research demonstrates that investment strategies optimized using the SS Ratio outperform those based on traditional metrics, showcasing its potential for superior returns [5].
• He has also made significant contributions in applying AI to stock selection for financial institutions. He has developed various models leveraging diverse approaches, including:
  • Text-mining sentiment analysis: This model analyzes textual data to identify stocks with promising sentiment, aiding investment decisions [6].  
  • Graph Laplacian-Based Multi-task Learning: This model exploits the interconnectedness of various stocks to recommend promising investment targets [7]. 
  • Predictor and historical probability-based performance: This model utilizes fundamental and technical indicators along with historical trends to identify potentially lucrative stocks [8]. Dr. Khushi's innovation extends beyond investment selection. 
  • developed proprietary algorithms for generating synthetic data [9], a valuable tool for various applications:  
  • Fraud detection: Identifying fraudulent activities within financial systems.  
  • Credit default prediction: Anticipating potential loan defaults for improved risk management [10]. 
  • Corporate bankruptcy: Predicting corporate bankruptcy before it occurs, mitigating financial risks [11]. His expertise expands even further to derivative markets. He has developed models for complex financial instruments like CFDs, Options and Futures, facilitating informed trading decisions in these markets [5, 12-14].  

Cited Publications of Dr. Khushi: 

1. Kim, T.W. and M. Khushi. Portfolio Optimization with 2D Relative-Attentional Gated Transformer. in 2020 IEEE Asia-Pacific Conference on Computer Science and Data Engineering (CSDE), Gold Coast, QLD, Australia, 16–18 December 2020. 2020. 
2. Huang, A., M. Khushi, and B. Suleiman, Regime-Specific Quant Generative Adversarial Network: A Conditional Generative Adversarial Network for Regime-Specific Deepfakes of Financial Time Series. Applied Sciences, 2023. 13(19): p. 10639. 
3. Wang, X., et al., Learning Non-Stationary Time-Series with Dynamic Pattern Extractions. IEEE Transactions on Artificial Intelligence, 2021. 
4. He, J., et al. Robust Dual Recurrent Neural Networks for Financial Time Series Prediction. in Proceedings of the 2021 SIAM International Conference on Data Mining (SDM). 2021. SIAM. 
5. Zhang, Z. and M. Khushi. GA-MSSR: Genetic Algorithm Maximizing Sharpe and Sterling Ratio Method for RoboTrading. in 2020 International Joint Conference on Neural Networks (IJCNN). 2020. 
6. Jaggi, M., et al., Text Mining of Stocktwits Data for Predicting Stock Prices. Applied System Innovation, 2021. 4(1): p. 13. 
7. He, J., N.H. Tran, and M. Khushi. Stock Predictor with Graph Laplacian-Based Multi-task Learning. in International Conference on Computational Science. 2022. Springer International Publishing Cham. 
8. Singh, J. and M. Khushi, Feature Learning for Stock Price Prediction Shows a Significant Role of Analyst Rating. Applied System Innovation, 2021. 4(1). 
9. Mukherjee, M. and M. Khushi, SMOTE-ENC: A Novel SMOTE-Based Method to Generate Synthetic Data for Nominal and Continuous Features. Applied System Innovation, 2021. 4(1): p. 18. 
10. Alam, T.M., et al., An Investigation of Credit Card Default Prediction in the Imbalanced Datasets. IEEE Access, 2020. 8: p. 201173-201198. 
11. Alam, T.M., et al., Corporate bankruptcy prediction: An approach towards better corporate world. The Computer Journal, 2021. 64(11): p. 1731-1746. 
12. Zhao, Y. and M. Khushi. Wavelet Denoised-ResNet CNN and LightGBM Method to Predict Forex Rate of Change. in 2020 IEEE International Conference on Data Mining Workshops (ICDMW), Sorrento, Italy, 11–17 November 2020. 2020. 
13. Zeng, Z. and M. Khushi. Wavelet Denoising and Attention-based RNN- ARIMA Model to Predict Forex Price. in 2020 International Joint Conference on Neural Networks (IJCNN). 2020. 
14. Qi, L., M. Khushi, and J. Poon. Event-Driven LSTM For Forex Price Prediction. in 2020 IEEE Asia-Pacific Conference on Computer Science and Data Engineering (CSDE), Gold Coast, QLD, Australia, 16–18 December 2020. 2020.