Research

  • Research

Trustworthy AI Service Systems

By leveraging his research activities built around service computing and smart services for the Internet of Things (IoT), Dr Badr current research strategy aims at developing new chain of data analytical models and tools and platforms for designing and deploying “Trustworthy AI Service Systems.” To this end, he is investigating a multidisciplinary and systematic approach, integrating AI, IoT and Blockchains and focusing on the following research areas:

  • AI-based Service Systems
  • Blockchains and AI-based Services
  • Trustworthiness and AI Systems
  •  
    1. AI-based Service Systems
    2. Blockchains and AI-based Services
    3. Trustworthiness and AI Systems

Keywords: Applied Machine Learning and Reinforcement Learning, Blockchains, Service Computing, IoT.

Recent Research Projects and Grants

Managing Risks in AI Systems: Mitigating Vulnerabilities and Threats Using Design Tactics and Patterns

      • Team:
        • Youakim Badr (PI), School of Graduate Professional Studies
        • Raghu Sangwan, Satish Srinivasan, and Partha Mukherjee, (School of Graduate Professional Studies) (co-PIs)
        • Prasenjit Mitra, College of IST, Technical Consultant
      • Program: 2020 industryXchange Grant
      • Budget: $48,000, Period: 10/01/2020 – 10/01/2021

Advances in AI combined with sensors, actuators and embedded systems technologies has made it feasible to incorporate intelligence into software intensive-systems with the ability to control and adapt their behavior in real time. Designing AI-centric systems, therefore, has become and will be a norm in the future. These systems are likely to be distributed Managing the complexity that comes with designing such dynamic systems requires risk management to handle uncertainty, safety and dependability that, if not addressed, can make these systems vulnerable to potential threats. However, cybersecurity and vulnerability of AI models to adversarial attacks have raised concerns and lead AI models to misclassify or misbehave. 

This research project has the following objectives:

  • Develop AI Risk Management Framework from holistic and multi-disciplinary perspectives to identify cyber threats and assess cyber-risks and mitigation strategies.
  • Develop fault tolerance mechanisms for AI models to ensure their resilience in production.
  • Extend software testability to AI testability and define new test tactics and patterns.
  • Develop monitoring mechanisms to detect propagation  of threats and vulnerabilities in distributed environments. 

Crowdlearning: Building Trustworthy AI Models from Crowdsourced Data and Edge Computing

      • Team:
        • Youakim Badr (PI), School of Graduate Professional Studies
        • Prasenjit Mitra (co-PI), College of Information Sciences and Technology
      • Program: Center for Security Research and Education – Impact Award
      • Budget: $57,467; Period: 01/11/2021 – 2/01/2022

This project introduces the concept of “crowdlearning” as a participatory method of building AI models, such Deep Neural Networks (DNN), with the help of a large group of contributors. In crowdlearning, contributors, as paid freelancers or volunteers, collaboratively participate to train a global AI model while keeping all their sample data on their devices (personal PC, mobile phones, self-driving cars, etc..). The project aims at developing a secure framework, called crowdlearning platform, with cluster federated learning to enable a collaborative train AI models from local data in trustless environments. The platform includes pruning algorithms to mitigate backdoors, and relies on blockchain-based cybersecurity mechanisms to control access to local data and edge devices and resist DDoS attacks. 

The project tackles the following challenges: 

  • Build a trustworthy federated learning algorithm resistance to adversarial attacks and backdoors.
  • Develop digital identity and authorization protocols without relying on external security authority (i.e., third party identity provider) to empower contributors with full controls on how their local data and computational resources are used. 
  • Develop a fully functional prototype will implement the crowdlearning platform, and demonstrate its feasibility and performance on real-world datasets, including Projected Healthcare Information under HIPAA.

Trusty AI: Trustworthy and reliable Federated Learning with privacy preserving

      • Team:

        • Antoine Boutet (PI), Jan Aalmoes and Thomas Lebrun (co-PIs) – INRIA / INSA de Lyon, France

        • Youakim Badr (co-PI), Robin Qiu, Prasenjit Mitra, Patrick McDenial – Pennsylvania State University 
      • Program: Pack Ambition International 2021- Auvergne-Rhône Alpes Region, France
      • Total awarded: $57,467; Period: 01/11/2021 – 2/01/2022

This project aims at developing a secure Federated Machine Learning Framework and tools that preserve the confidentiality of personal data in distributed environments. To this end, we will extend different federated learning approaches and consider their limitations in terms of accuracy, confidentiality, and robustness related to these approaches. In addition, we will enable our Federated Machine Learning Framework with mechanisms to better understand the distributed AI learning process and ensure unbiased fairness that may occur from users data.

This project will also strengthen a partnership between professors from INSA Lyon, School of Graduate Studies, College of Information Science and Technologies, and School of Computer Science and  Engineering at the Pennsylvania State University to not only develop common research topics but also exchange Ph.D. students in Lyon and Penn State research laboratories. 

In addition, the project aims to develop several teaching initiatives to allow students from both institutions to benefit from the Federated Machine Learning Framework as a teaching platform to build federatel learning projects and experiments Cybersecurity attacks on  AI systems. The mutual visits of faculty also aim to promote double degree programs and summer programs.

Research Projects (Assistantship and graduate students)

AI-based Service Systems

The key challenge under this topic revolves around new analytic methods and techniques to improve human-to-machine cognitive-based services in the Cloud or at the edge.

1) General Purpose Conversional AI Expert (Impulso)
      • Team: Haruka George’22(MS), Atharva Mungee’21 (MS) and Dr. Robin Qiu
      • Objective: build a Conversational AI service based on cognitive processes and learning objectives in a closed domain. The Conversational AI service continuously adapts conversions and answers based on accumulated knowledge acquired from interactions with humans and the assessment of their cognitive capabilities.
2) Non-Verbal Behavior Analyzer (NOVOR)
        • Team: Ambika Chundru’22, Shraddha Maurya’19 (RA), Sura Bondugula (RA), and Dr. Minyoung Cheong
        • Objective: develop a platform and models to detect patterns of non-verbal behaviors when humans interact with each other’s and/or with virtual assistants.

Blockchains and AI-based Services

The integration of Blockchains and AI still a largely undiscovered area and their combination have the potential to build new services in ways never before thought possible. To tackle this challenge, I am leading complementary research projects, covering AI for Blockchains, Blockchains for AI and distributed applications based on the integration of AI and Blockchains.

1) Integration of AI and Blockchains 3.0 (chAIns)

      • Team: Harsh Deokuliar ’22, Vineeth Suhas Challagali’20 (RA) and Dr. Partha Mukherjee
      • Objective: identify solutions to cope with challenges of existing blockchains in terms of interoperability, scalability, privacy and computational capabilities when using them to build end-to-end machine learning systems.

2) AI for Blockchains 

2.1)  Blockchain Data Analytics (Daan.chains)
          • Team: Akash Singh Baghel’20 (RA) and Dr. Partha Mukherjee
          • Objective: build analytics pipelines to explore, understand and get insights from Ethereum and Bitcoin blockchains. 
2.1) Cryptocurrencies Exchange Rates Forecasting
          •  Team: Gauravi Bhalchandra Patil’20 (RA), Mokkapati, Yogitha Siva’20 (RA) and Dr. Partha Mukherjee
          •  Objective: deploy state of the art Deep Neural Networks models to forecast the Ethereum and Bitcoin Exchange Rates.

Trustworthiness and AI Systems

Recent advances in AI outperform in many cognitive tasks and become omnipresent in decision-making systems, self-driving cars, and critical systems. AI systems also present risks and biases and we must carefully consider their safety, trustworthiness, robustness, and dependability. In this topic, I am interested in AI risk management to identify vulnerability, assess, and mitigate risks at design and deployment time.

1) Trustworthy Federated Learning 

        • Team: Sameer Arun Mahajan’22 (MS), RVirajdatt Vishaldatt Kohir’22 (RA), Rahu Sharma’20 (RA) and Suraj Bondugula’20 (RA)
        • Objective: Build an automated tool to detect adversarial attacks and biases in Federated Learning

2) Auto-Adversarial and Bias Vulnerability Detection

        • Team: Rahu Sharma’20 (RA) and Suraj Bondugula’20 (RA)
        • Objective: Build an automated tool to detect adversarial attacks and biases.

3) Smart and Secure Devices

        • Reinforcement Learning-based energy consumption Controller
          • Team: Anchal Gupta’19 (MS), Dr. Robin Qiu, Dr. Ashkan Negahban
          • Objective: build and deploy Reinforcement Learning Controller on devices
        • Reinforcement Learning-based-Intrusion detection at the edge
          • Team: Wahid Khan Abzal’19 (RA)
          • Objective: build and deploy an intrusion detector at the edge (Raspberry Pi)
  • Legends:
    RA – Research Assistant
    MS – MS graduate student

Past Projects

 Cybersecurity Collaboratory: Cyberspace Threat Identification, Analysis & Proactive Response (2013-2018)

Source of Support: Partner University Fund (PUF) (USA-France)

Academic Partners: University of Lyon 1 (LIRIS Lab), University of Arizona and University of Chicago

Keywords: Information Assurance, Security by Design, Moving Target Techniques, Resilient SOA, etc. 

 

RIOT: Resilient Security in Dynamically Networked Smart Object (IoT) (2015-2016)

Source of Support: Seed Grant, INSA-Lyon’s inter-labs funding program

   Partners: LIRIS Lab (INSA-Lyon/University of Lyon 1)

Keywords: Mobile devices, Access control, Delay Tolerant Networking, Simulation, …

 

LLIOT: Linear Logic for the Internet of Things (2018-2020)

Source of Support: Informatic Federation of Lyon Grant

Partners: LIRIS Lab (INSA Lyon), LIP (ENS Lyon)

Keywords: Linear Logic, automated prover, proof certification, Coq, OCaml.

 

Brain 2.0: Brain-Smart Object Interfaces for the Elderly People to Control Home Devices with Brainwaves (2016-2018)
Source of Support: COOPERA-International Collaboration Program, AURA

      Project life span: 2016-2018  

Partner: The University of Pittsburgh, LIRIS lab (INSA)  

      Keywords: Internet of Things, formal specification, event streaming, signal processing, etc.


CT-ANALYTICS: Big Data Proactive Analytics Platform for Analyzing Citizen behaviors in Urban Worlds (2015-2017)

Source of Support: COOPERA-International Collaboration Program, AURA

Partner: The Pennsylvania State University, LIRIS Lab (INSA-Lyon)

Keywords: Big Data Analytics, Open data, sentiment Analysis, User behavior Analysis, etc.

 

SemEUse: Design of Semantic and Secure Enterprise Service Bus (2008-2010)

       Source of Support: French National Research Agency (ANR) 

 Industrial Partners: Thales Communication France, France Télécom, EBM Websourcing,

 Academic Partners: INRIA Object Web, INRIA ARLES, Télécom SudParis, LIP6, LIESP Lab (INSA-Lyon)

 Keywords: Security, Late Binding, Monitoring, QoS, SOA, ESB, Ontology, etc.

 

 ISPRI-PLM: Services for the Integration of Industrial Processes and their Application to Product Lifecycle Management (2009-2011)

      Source of Support: Rhône-Alpes Region

Academic Partners: G-SCOP, IREGE, LIESP Lab (INSA-Lyon), LISTIC, STOICA, SYMME, INRIA, SYSCOM

Industrial Partners: ASSETIUM, Arve Industries, EBM Websourcing, AIP-Primeca Rhône Ouest

Keywords: Product Life-Cycle, SOA, Interoperability, Model Driven Engineering, Standardization.

 

INTER-PROD: Organizational & Technological Interoperability to Support Co-Production (2006-2008)

Source of Support: Rhône-Alpes Region

Academic Partners: G2I, LIESP Lab (INSA-Lyon)

Industrial Partners: EBM Websourcing

Keywords: Co-production, Enterprise Architecture, Organizational Structure, Collaboration, … 

 

COPILOTES: Collaboration and Information Exchange in Supply Chains (2004-2006)

Project life span: 

Academic Partners: PRISMa /INSA de Lyon/Lyon 2, COPISORG, G.A.E.L./Grenoble, GILCO/ (Institut National Polytechnique de Grenoble, ENSMSE_G2I/ ENS des Mines de Saint-Etienne.

Industrial Partners: RHODIA, ROSET, FINMATICA, VALRHONA

Keywords: Value Network, Supply Chain, Information Sharing, Process Integration, Best Practice …

 

BSM: Business Collaborative Service Bus (2009-2011)

Source of Support: Seed grant (INSA-Lyon)

Partners: LIRIS Lab (Database research team, Distributed System research team, SOC research team

Keywords: ESB, Web services, service composition, business processes, …

 

PERS: Service-Based Pervasive Environment to Assist Elderly People (2005-2007)

 Partners: LIESP Lab (INSA-Lyon), LIRIS, EMPERE (INSA-Lyon Inter Labs research project)

 Source of Support: Seed Grant, INSA-Lyon’s inter-labs funding program

 Keywords: ubiquitous computing, sensors, service composition, machine learning, …