About Dr Matloob Khushi & MatFinProtocol

Who am I?

I am Dr Matloob Khushi, Associate Professor and Senior Lecturer in Artificial Intelligence at Brunel University London, and Deputy Director of the Doctoral Programme in the Department of Computer Science. I hold a PhD in Bioinformatics and Data Science from the University of Sydney, and my work has been recognised among the world’s top 2% of scientists (Stanford list, 2024/2025) with over 4,000 citations across AI, finance and computational biology.

My research spans AI in financial markets (algorithmic trading, stock prediction, risk‑adjusted optimisation), bioinformatics and computational biology (genomic interactions, diagnostics, drug discovery), and natural language processing. I regularly publish in venues such as IEEE Transactions and PeerJ, and I create educational content on AI and financial markets across TikTok, YouTube and podcast platforms.

What do I research?

On the finance side, my group focuses on turning machine‑learning predictions into deployable, risk‑controlled trading strategies. This includes:

• Designing and stress‑testing AI models for equity, FX, crypto and multi‑asset trading.
• Developing risk‑adjusted performance measures (e.g. the SS Ratio) that jointly account for volatility and drawdowns.
• Formalising anti‑leakage, backtesting and reporting standards for financial ML — the core of MatFinProtocol.

In bioinformatics and medical AI, I work on predictive models for genomic interactions and clinical outcomes, with an emphasis on reproducibility and rigorous evaluation — the same mindset that underpins this leaderboard.

Who manages this page?

This leaderboard is curated by Dr Matloob Khushi and collaborators, and is intended as a living, protocol‑compliant register of AI‑based financial forecasting and trading studies. Only backtests that meet the MatFinProtocol criteria (no data leakage, realistic costs, fully specified trading rules, and comprehensive risk reporting) are included.

The goal is not to “pick winners” in a vacuum, but to provide a transparent, comparable view of what has actually been demonstrated in the literature under consistent assumptions, so that researchers, practitioners and investors can make informed decisions.

Contact & reporting errors

If you spot any errors or omissions, or if you would like to suggest a new protocol‑compliant study to add to the leaderboard, please get in touch. I welcome high‑quality contributions and critical feedback from both academia and industry.