[PhD] Hybrid Neural Networks for Anomaly Detection in Cyber Physical Systems

Keywords : Artificial Intelligence, Anomaly Detection, Physics guided machine learning

Contact: 

Abdelkader LAHMADI (lahmadi@loria.fr),

Jérôme François (Jerome.francois@inria.fr),

Isabelle Chrisment (isabelle.chrisment@loria.fr)

 

Scientific Context:

The security monitoring Cyber Physical Systems (CPS) is a hot research topic since they are used in multiple application domains ranging from smart homes, smart cities, hospitals, to critical infrastructures, etc. These systems mainly rely on different wireless communication protocols (BLE, Zwave, Wifi, Zigbee, LORA, etc) and have different deployment models (device to cloud, device to device and device to gateway). They usually couple two entities: a computing and networking component that represents the cyber part of the system and physical components that is controlled and sensed to meet the application requirements.

Nowadays, CPS environments are also faced to multiple cyber-threats. Being exposed to cyber specific threats, their malfunctioning can seriously disconnect the computation from the control entity of a device which can lead to severe impacts on the real world and the safety of people like grid electricity blackout or water contamination. Different algorithms and methods have been designed for the security monitoring of such systems with different goals including anomaly detection and intrusion detection.

Recently, machine learning and deep learning algorithms are applied to detect such anomalies and attacks. However, most of the applied methods only rely on the cyber part of these systems and on the data that describe their behavior without considering their physical models. In this internship, we investigate how to employ hybrid machine learning technique, in particular to apply neural networks to detect anomalies in CPS while considering its physical model.

Subject and goals:

 The first objective of this PhD is to elaborate a state of the art regarding Physics guided or hybrid neural networks [1,2] and their domain applications. In a second task, the candidate will adapt and extend machine learning algorithms including neural networks to consider a physical model of a CPS system such a microgrid or a process control system to detect anomalies including attacks and faults. In a third task, the candidate validate the designed method and evaluate its performance compared to traditional methods.

References:

[1] Xiaowei Jia, Anuj Karpatne, Jared Willard, Michael Steinbach, Jordan Read, Paul C Hanson, Hilary A Dugan, Vipin Kumar, “Physics Guided Recurrent Neural Networks For Modeling Dynamical Systems: Application to Monitoring Water Temperature And Quality In Lakes”, 8th International Workshop on Climate Informatics.  https://arxiv.org/abs/1810.02880

[2] A. Karpatne, “How can Physics Inform Deep Learning Methods in Scientific Problems?: Recent Progress and Future Prospects,” Workshop on Physics Informed Machine Learning, 2018.

RESIST team:

The PhD position is proposed by the RESIST team of the Inria Nancy Grand Est research lab, the French national public institute dedicated to research in digital Science and technology.  The team is one of the European research group in network management and is particularly focused on empowering scalability and security of networked systems through a strong coupling between monitoring, analytics and network orchestration. https://team.inria.fr/resist/

Skills and profile:

  • Required qualification: Master in computer science
  • Knowledge and skills in the following fields will be appreciated: networking, security,
  • machine learning, big data, programming (C, python)

Additional information:

Duration: 3 years

Starting date: between Oct. 1st 2020 and Jan. 1st 2023

How to apply:

Send the following documents to jerome.francois@inria.fr and abldekader.lahmadi@loria.fr:

– CV;

– a motivation letter;

– your degree certificates and transcripts for Bachelor and Master (or the last 5 years if not applicable).

– Master thesis (or equivalent) if it is already completed, or a description of the work in progress, otherwise;

– all your publications, if any (it is not expected that you have any).

– At least one recommendation letter from the person who supervises(d) your Master thesis (or research project or internship); you can also send at most two other recommendation letters. The recommendation letter(s) should be sent directly by their author to the prospective PhD advisor.

Logo d'Inria