Monthly Archives: October 2021

PROFESSOR OR ASSISTANT/ASSOCIATE PROFESSOR IN SOFTWARE ENGINEERING

|

The Department of Computer Science at the Faculty of Science of the University of Helsinki invites applications for a

PROFESSOR OR ASSISTANT/ASSOCIATE PROFESSOR IN SOFTWARE ENGINEERING

aiming to strengthen research areas in computer science at the university.

Position description
We are looking for a new professor or assistant/associate professor to carry out research into the field of software engineering.

The person to be chosen must have a strong scientific track record in the field of software engineering evidenced by publications in top tier forums of the field. The University of Helsinki and the Department of Computer Science offer excellent potential collaborators in the relevant areas, such as data science and other disciplines of different faculties of the university.

The appointee is expected to conduct research on an internationally high level. We expect a strong network of international and interdisciplinary cooperation and collaboration. In addition, strong record in empirical software engineering and particularly in close industrial collaboration is considered a significant advantage. The teaching area of the position is software engineering. The person chosen for the position is expected to teach and develop teaching particularly in the Master’s Programme in Computer Science and potentially also in the Bachelor Programme in Computer Science and the Bachelor’s Programme in Science. In addition, the appointee will participate in doctoral education at the Department of Computer Science, collaborate with other groups in the department, acquire research funding for their research group, and participate in the interaction between science and society at large.

More information: https://www2.helsinki.fi/en/open-positions/professor-or-assistantassociate-professor-in-software-engineering

MLOps research and thesis positions

|

Machine learning (ML) is becoming an increasingly common technology that is used in software-based systems. While software and software engineering practices are still a quintessential part of the design of such systems, the design space has data and ML and respective data scientist stakeholders as new concerns. Hence, MLOps — as a derivative of DevOps — is currently emerging to better integrate data and ML into the software engineering way of working, as well as integrate ML better in the software-based systems. MLOps is about both practices and tools for ML-based systems that technically enable iterative software engineering practice.

The ESE research group is strengthening its research focus in the area of MLOps. In particular, we are participating in projects that study MLOps, such as AIGA https://ai-governance.eu/ and recently started IMLE4 https://itea4.org/project/iml4e.html.

We have funded thesis positions in the area of MLOps in these research projects that can be tailored to the interest of the applicant. For details, contact Mikko Raatikainen (first.last@helsinki.fi).

Misbehaviour and fault tolerance in machine learning systems

|

Our interview paper on misbehaviour and fault tolerance in machine learning systems was recently published in Journal of Systems and Software. Machine learning (ML) provides us with numerous opportunities, allowing ML systems to adapt to new situations and contexts. At the same time, this adaptability raises uncertainties concerning the run-time product quality or dependability, such as reliability and security, of these systems. Systems can be tested and monitored, but this does not provide protection against faults and failures in adapted ML systems themselves.

We studied software designs that aim at introducing fault tolerance in ML systems so that possible problems in ML components of the systems can be avoided. The research was conducted as a case study, and its data was collected through five semi-structured interviews with experienced software architects.

We present a conceptualisation of the misbehaviour of ML systems, the perceived role of fault tolerance, and the designs used. The problems in the systems rise from problems in inputs, concept drift, bugs and inaccuracies in the models, their faulty deployment, and not really understanding what the utilised model does. Common patterns to incorporating ML components in design in a fault tolerant fashion have started to emerge. ML models are, for example, guarded by monitoring the inputs and their distribution, and enforcing business rules on acceptable outputs. Multiple, specialised ML models are used to adapt to the variations and changes in the surrounding world, and simpler fall-over techniques like default outputs are put in place to have systems up and running in the face of problems.

However, the general role of these patterns is not widely acknowledged. This is mainly due to the relative immaturity of using ML as part of a complete software system: the field still lacks established frameworks and practices beyond training to implement, operate, and maintain the software that utilises ML. ML software engineering needs further analysis and development on all fronts.

The full paper can be read here (open access): https://doi.org/10.1016/j.jss.2021.111096