Dear Colleagues,

There has been growing interest regarding integrating machine learning (ML) into a wide range of software systems, from digital assistants to self-driving software. While the ML research community is rapidly progressing state-of-the-art ML technologies, such as deep learning, vision, speech, and NLP, building software systems that include these core ML technologies (also known as ML systems) can be challenging.

Specifically, developing ML systems requires compliance with the complex ML workflow, which includes several stages, such as ML model requirements, data collection, data cleaning, data labeling, feature engineering, model training, model evaluation, model deployment, and model monitoring. Because of this complexity, traditional software engineering (SE) practices need to evolve to support the effective development of ML systems.

Furthermore, ML systems are highly data-dependent and often prone to model drift (or decay), caused by changes in the model environment. There are two types of model drift: concept drift, where the joint distribution of the model’s inputs and outputs changes, and data drift, where the distribution of the model’s input changes. Both types, if not detected and handled properly, can significantly decrease the ML model’s prediction performance.

The purpose of this Special Issue is to present advancements in the area of software engineering for the development of machine learning systems.

Topics of interest include, but are not limited to, the following:

• Methods for supporting ML system engineering at different stages of the ML workflow (model requirements, data collection, data cleaning, data labeling, feature engineering, model training, model evaluation, model deployment, and model monitoring).
• Requirement engineering for ML systems.
• Verification and validation methods for ML systems.
• Empirically validated studies of applying SE methods for the development of ML systems.
• Systematic reviews of the state-of-the-art and state of practice in software engineering for designing, developing, deploying, operating, maintaining, and evolving ML systems.

Dr. Dusica Marijan
Guest Editor

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Applied Sciences is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• software engineering for machine learning: techniques, methods, and tools
• autonomous systems
• self-managing systems
• socio-technical systems
• human aspects of software engineering for machine learning
• trust and trustworthiness
• artificial intelligence systems
• artificial intelligence system engineering
• human-centered AI