*SIAM Journal on Mathematics of Data Science* (SIMODS) publishes work that advances mathematical, statistical, and computational methods in the context of data and information sciences. We invite papers that present significant advances in this context, including applications to science, engineering, business, and medicine.

Topics of interest include but are not limited to:

- numerical algorithms
- statistical inference
- optimization and control
- machine learning
- theoretical computer science
- signal processing and information theory
- applied probability
- functional analysis
- network science

Application papers should contain a new methodology beyond the specific application in areas such as astrophysics, chemistry, climate, cosmology, earth science, materials science, bioscience, biomedicine, pharmaceutical, biophysical modeling, neuroscience, economics, manufacturing, transportation, banking, finance, security, privacy, social and behavioral science, internet of things, and cyber science.

A SIMODS paper will make a novel contribution in one or more of the following ways:

- Advances in theoretical analysis that motivate new approaches, develop accuracy bounds, or otherwise provide guidance on the practical computational usage of methods.
- Novel algorithms and/or implementations that provably and significantly improve on the state-of-the-art for mathematical data analysis in terms of accuracy, running time, memory efficiency, scalability, or other computational measures. This includes adapting existing algorithms or approaches to exploit contemporary computing architectures, new statistical paradigms, or heterogeneous information sources.
- The solution of a specific practical data science problem that uses novel methodology and is of broad interest. The description should contain enough information about the application to orient other data scientists but should omit details of interest mainly to the applications specialist.

All papers must be clearly written and understandable to a broad audience. The novel contributions should be clearly called out in the abstract and introduction. Mathematical notation and terminology should be as clear as possible, with enough details to be understandable by a wide audience. We especially value papers that explain both the pros and cons of newly proposed methods. Experimental results should focus on problems that are representative of real-world applications.

Authors should include sufficient information in the manuscript to enable the results of the manuscript to be reproduced. To this end, authors should deposit in a permanent repository, or as supplementary materials, relevant software and data, and should include in the manuscript or supplementary materials information used for data preparation, a precise description of the methods, choices for all parameters, all testing conditions, and details on post-processing to recover published results.

Brevity is encouraged, with a suggested maximum length of 20 pages. Longer papers are published only in exceptional cases. In such cases, the editor and referees must be convinced that the length of the paper is both required by the subject matter and justified by the paper’s quality.

All submissions must adhere to journal publication policies.