Curious About AI
From NORDUnet
views
comments
Related Media
Explore the development of HAWAT, an agentic AI assistant designed for network troubleshooting. This innovative system leverages advanced technologies like Large Language Models (LLMs), Retrieval-Augmented Generation (RAG), and Reasoning and Acting (ReAct) frameworks to autonomously manage network conditions and provide natural language interfaces for network administrators. Next, dive into the AI4LAM initiative, which is revolutionizing the cultural heritage sector. This collaborative network is dedicated to advancing AI tools and services for libraries, archives, and museums, enhancing the management and accessibility of digitized content while fostering innovation and knowledge sharing. Finally, discover the FedXAI4DNS project, which employs Federated Learning and Explainable AI to bolster DNS security in privacy-aware environments. This project showcases how AI can collaboratively detect malicious traffic, ensuring network security without compromising user privacy. This session promises to be a captivating journey through the latest AI innovations, offering valuable insights for anyone curious about the transformative potential of artificial intelligence.
Chair: Chris de loof (Belnet)
HAWAT: An Agentic AI Assistant for Network Troubleshooting
HAWAT (Heuristic Analysis With Adaptive Troubleshooting) is an agentic AI assistant for network troubleshooting. This presentation explores the development of a chatbot system designed to interface with network hardware, leveraging recent advances in Large Language Models (LLMs), Retrieval-Augmented Generation (RAG), and Reasoning and Acting (ReAct) agentic frameworks. Our system demonstrates how agentic AI can autonomously interpret network conditions, execute commands, and provide network administrators with natural language interfaces to network infrastructure. We will detail how the system was built and present pre recorded demos of the chatbot in action.
Speakers: Karl Newell (Internet2), Matt Mullins (Internet2)
HAWAT (Heuristic Analysis With Adaptive Troubleshooting) is an agentic AI assistant for network troubleshooting. This presentation explores the development of a chatbot system designed to interface with network hardware, leveraging recent advances in Large Language Models (LLMs), Retrieval-Augmented Generation (RAG), and Reasoning and Acting (ReAct) agentic frameworks. Our system demonstrates how agentic AI can autonomously interpret network conditions, execute commands, and provide network administrators with natural language interfaces to network infrastructure. We will detail how the system was built and present pre recorded demos of the chatbot in action.
Speakers: Karl Newell (Internet2), Matt Mullins (Internet2)
AI4LAM: Advancing AI in Libraries, Archives, and Museums through Collaboration with the Geant Community
The AI for Libraries, Archives, and Museums (AI4LAM) community is an international, participatory network dedicated to advancing the use of artificial intelligence within the cultural heritage sector. The community is at the forefront of developing and maintaining cutting-edge AI tools and services tailored for heritage institutions to better provide access, management and (re)use of digitized and digitally born content by supporting collaboration, innovation, and sharing of knowledge in the field of AI for institutions worldwide.
AI4LAM's commitment to openly share and provide access to knowledge aligns with Open Science principles to make scientific research more transparent, accessible, and collaborative. The partnership between AI4LAM and the GÉANT community represents a unique opportunity to advance the use of AI in LAM. By working together, we can build a more innovative, secure, and collaborative future.
Speaker: Ines Vodopivec (Assist. prof. dr. Ines Vodopivec (presenter), Secretary General of AI4LAM, Europeana Network Association Management Board Member)
The AI for Libraries, Archives, and Museums (AI4LAM) community is an international, participatory network dedicated to advancing the use of artificial intelligence within the cultural heritage sector. The community is at the forefront of developing and maintaining cutting-edge AI tools and services tailored for heritage institutions to better provide access, management and (re)use of digitized and digitally born content by supporting collaboration, innovation, and sharing of knowledge in the field of AI for institutions worldwide.
AI4LAM's commitment to openly share and provide access to knowledge aligns with Open Science principles to make scientific research more transparent, accessible, and collaborative. The partnership between AI4LAM and the GÉANT community represents a unique opportunity to advance the use of AI in LAM. By working together, we can build a more innovative, secure, and collaborative future.
Speaker: Ines Vodopivec (Assist. prof. dr. Ines Vodopivec (presenter), Secretary General of AI4LAM, Europeana Network Association Management Board Member)
FedXAI4DNS: Explainable AI for DNS Security in Privacy-aware NREN Federations
Machine Learning (ML) has seen limited adoption within large-scale networks (e.g. NRENs). Organisations are reluctant to share their data in fear of compromising end-user privacy, thus representative datasets to train accurate ML classifiers are usually not available. Moreover, complex black-box ML classifiers are not intrinsically explainable, hence network engineers are reluctant to deploy them. We present FedXAI4DNS that employs ML, Federated Learning (FL) and eXplainable AI (XAI) for collaborative and trustworthy detection of malignant DNS traffic produced by Domain Generation Algorithms (DGAs). FL enables collaborating organisations to jointly train privacy-aware classifiers without exchanging sensitive data, whereas XAI suggests methods for justifying configurations of complex black-box models. FedXAI4DNS aims at expediting ML adoption within collaborative environments (e.g. NRENs & GÉANT).
Speakers: Maria Grammatikou (ICCS/NTUA), Mr Nikos Bazotis (National Technical University of Athens, NTUA / ICCS)
Machine Learning (ML) has seen limited adoption within large-scale networks (e.g. NRENs). Organisations are reluctant to share their data in fear of compromising end-user privacy, thus representative datasets to train accurate ML classifiers are usually not available. Moreover, complex black-box ML classifiers are not intrinsically explainable, hence network engineers are reluctant to deploy them. We present FedXAI4DNS that employs ML, Federated Learning (FL) and eXplainable AI (XAI) for collaborative and trustworthy detection of malignant DNS traffic produced by Domain Generation Algorithms (DGAs). FL enables collaborating organisations to jointly train privacy-aware classifiers without exchanging sensitive data, whereas XAI suggests methods for justifying configurations of complex black-box models. FedXAI4DNS aims at expediting ML adoption within collaborative environments (e.g. NRENs & GÉANT).
Speakers: Maria Grammatikou (ICCS/NTUA), Mr Nikos Bazotis (National Technical University of Athens, NTUA / ICCS)
- Tags
-