Morgan Stanley Machine Learning Research

This respository contains code for papers and projects published by the Morgan Stanley Machine Learning Research team.

Who We Are

The Machine Learning Research team at Morgan Stanley harnesses the advances of machine learning techniques across the firm. The ML Research team is comprised of hyper-specialist researchers who work on fundamental and complex problems. The team has worked in a wide variety of areas including time series analysis, recommender systems, network theory, NLP for finance, fairness, privacy, and more. The central ML Research team spearheads engagements with academia, both by collaborating with university labs and by publishing in top-tier conferences.

Papers

See also: https://www.morganstanley.com/about-us/technology/machine-learning-research-papers

Year	Venue	Title	Code
2025	ICML	Privacy Amplification by Structured Subsampling for Deep Differentially Private Time Series Forecasting	code
2025	TMLR	Reweighting Improves Conditional Risk Bounds	code
2025	AISTATS	Variational Schrödinger Momentum Diffusion	code
2024	UAI	Reflected Schrodinger Bridge for Constrained Generative Modeling	code
2024	NeurIPS	Recurrent Interpolants for Probabilistic Time Series Prediction	code
2024	NeurIPS	Efficient and Sharp Off-policy Evaluation in Robust Markov Decision Processes	code
2024	ICML	Variational Schrodinger Diffusion Models	code
2024	ICML	Constrained Exploration via Reflected Replica Exchange Stochastic Gradient Langevin Dynamics	code
2024	AISTATS	Neural McKean-Vlasov Processes: Distributional Dependence in Diffusion Processes	code
2024	AISTATS	Accelerating Approximate Thompson Sampling with Underdamped Langevin Monte Carlo	code
2023	UAI	Short-Term Temporal Dependency Detection Under Heterogeneous Event Dynamic With Hawkes Processes	code
2023	UAI	Information Theoretic Clustering via Divergence Maximization Among Cluster	code
2023	UAI	Inference and Sampling of Point Processes from Diffusion Excursions	code
2023	UAI	In- or Out-of-Distribution Detection via Dual Divergence Estimation	code
2023	TMLR	Learning to Abstain From Uninformative Data	code
2023	ICML	Provably Convergent Schrödinger Bridge with Applications to Probabilistic Time Series Imputation	code
2023	ICML	Modeling Temporal Data as Continuous Functions with Stochastic Process Diffusion	code
2023	ICLR Workshop	On the Existence of a Trojaned Twin Model	code
2023	AISTATS	Risk Bounds on Aleatoric Uncertainty Recovery	code
2022	UAI	Estimating Transfer Entropy Under Long-Ranged Dependencies	code

Projects

Name	Description	Links
qqWen	Qwen-2.5 series models finetuned for the Q programming language	HuggingFace,code
AlphaLab	Autonomous multi-agent research system across optimization domains with frontier LLMs	code

Contact

The team can be reached at msml-qa@morganstanley.com

Licensing

All files in this repository, unless explicitly mentioned otherwise, are released under the Apache 2.0 license, the text of which can be found in the LICENSE file.