BEGIN:VCALENDAR VERSION:2.0 PRODID:-//CERN//INDICO//EN BEGIN:VEVENT SUMMARY:SmartAttica Τraining modules for SMEs - Module 12 "Inference with Transformers & Semantic Search with Sentence Transformers" DTSTART;VALUE=DATE-TIME:20260622T080000Z DTEND;VALUE=DATE-TIME:20260622T110000Z DTSTAMP;VALUE=DATE-TIME:20260716T014300Z UID:indico-event-209@events.grnet.gr DESCRIPTION:\n\n\n\nGRNET announces\, in the context of SmartAttica EDIH ( European Digital Innovation Hub)\, the 12th Module of Τraining modules fo r SMEs with the subject "Inference with Transformers & Semantic Search wi th Sentence Transformers". \n\nDate: June 22nd\, 2026\n\nLocation: Onlin e via Zoom \n\nPresentation Languages: Greek\, English\n\nInstructors: Ni kos Bakas (GRNET)\, Roman Dolgopolyi (GRNET)\n\nDuration: 3 hours\n\nDescr iption: This is a hands-on introduction to running open Large Language Mod els and turning text into meaning with embeddings. Participants learn how to load a pre-trained model from the HuggingFace Hub\, generate text with streaming\, and control the tokenizer and chat template. The second half c overs Sentence Transformers\, showing how text is mapped into vectors and how cosine similarity reveals semantic relationships between words and sen tences - the foundation of modern search and retrieval systems.\n\nTarget Audience: This module is designed for SME developers\, technical leads\, a nd data scientists who want to incorporate natural language processing (NL P) into their projects. It is ideal for those looking to run models in the ir own environment and understand the building blocks behind semantic sear ch and RAG.\n\nLearning Objectives:\n\n By the end of this module\, parti cipants will be able to:\n\n\n Load and run pre-trained language models fr om the HuggingFace Hub using the Transformers library.\n Generate text wit h both streaming and standard inference\, and use the pipeline abstraction .\n Apply chat templates and manage tokenizers\, padding\, and special tok ens correctly.\n Produce embeddings from text using Sentence Transformers. \n Measure semantic similarity with cosine similarity and interpret the re sulting vector space. \n\n\nPrerequisites:\n\nParticipants should have:\n \n\n Basic understanding of Python programming.\n Familiarity with running code in Jupyter/Colab notebooks.\n Interest in NLP applications.\n Some e xperience with machine learning will be helpful.\n\n\nIndicative Content:\ n\n\n The Transformers Library. Introduction to HuggingFace and the ecosys tem for loading and running open models.\n Model and Tokenizer Setup. Down loading a model from the Hub\, configuring the tokenizer\, and handling pa d/EOS tokens.\n Chat Templates and Messages. Structuring system and user m essages and applying the model's chat template.\n Streaming Inference. Gen erating text token-by-token with a streamer for a responsive experience.\n Standard Inference and the Pipeline. Running batch generation and using t he high-level pipeline object.\n Introduction to Embeddings. What embeddin gs are and why similar meanings map to nearby vectors.\n Sentence Transfor mers in Practice. Encoding words and sentences into vectors with a compact \, fast model.\n Measuring Similarity. Using cosine similarity to compare texts and visualize a similarity matrix.\n Summary and Q&A. Key takeaways and open discussion.\n\n\n \n\nThe project is co-funded by the European U nion. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or the Europea n Commission. Neither the European Union nor the granting authority can be held responsible for them. \nhttps://events.grnet.gr/event/209/ LOCATION: URL:https://events.grnet.gr/event/209/ END:VEVENT BEGIN:VEVENT SUMMARY:SmartAttica Τraining modules for SMEs - Module 13 "Retrieval-Augm ented Generation (RAG) with Docling Document Parsing" DTSTART;VALUE=DATE-TIME:20260624T080000Z DTEND;VALUE=DATE-TIME:20260624T110000Z DTSTAMP;VALUE=DATE-TIME:20260716T014300Z UID:indico-event-210@events.grnet.gr DESCRIPTION:\n\n\n\nGRNET announces\, in the context of SmartAttica EDIH ( European Digital Innovation Hub)\, the 13th Module of Τraining modules fo r SMEs with the subject "Retrieval-Augmented Generation (RAG) with Doclin g Document Parsing". \n\nDate: June 24th\, 2026\n\nLocation: Online via Zoom \n\nPresentation Languages: Greek\, English\n\nInstructors: Nikos Ba kas (GRNET)\, Roman Dolgopolyi (GRNET)\n\nDuration: 3 hours\n\nDescription : This is a hands-on introduction to turning real-world documents into kno wledge a language model can use. Real documents are not plain text - they contain headings\, tables\, figures\, captions\, and reading order. Partic ipants learn how Docling converts a PDF into structured content using layo ut-aware vision models\, how to inspect and chunk that content intelligent ly\, and how to assemble a complete Retrieval-Augmented Generation (RAG) p ipeline: chunk\, embed\, retrieve\, and generate a grounded\, citable answ er.\n\nTarget Audience: This module is designed for SME developers\, techn ical leads\, and data scientists who want to build question-answering syst ems over their own documents. It is ideal for those looking to ground lang uage models in trusted internal knowledge sources.\n\nLearning Objectives: \n\nBy the end of this module\, participants will be able to:\n\n\n Conver t PDFs into structured documents with Docling and inspect the resulting do cument tree.\n Extract and work with text blocks\, tables\, figures\, and captions.\n Split documents into retrieval-ready chunks while preserving s tructure and context.\n Embed chunks and retrieve the most relevant ones u sing semantic similarity.\n Assemble a RAG prompt and generate an answer t hat cites its sources.\n\n\nPrerequisites:\n\nParticipants should have:\n\ n\n Basic understanding of Python programming.\n Familiarity with embeddin gs and semantic similarity.\n Interest in NLP and document-processing appl ications.\n Some experience with machine learning will be helpful.\n\n\nIn dicative Content:\n\n\n Why Document Structure Matters. The gap between ra w PDF text and structured\, machine-usable content.\n Converting PDFs with Docling. Layout detection\, reading order\, and the structured document t ree.\n Inspecting the Document. Exploring texts\, groups\, figures\, and c aptions\; previewing as Markdown.\n Working with Tables. Recovering rows\, columns\, and headers and exporting tables to Markdown.\n Chunking for RA G. Structure-aware splitting that keeps headings and context attached to e ach chunk.\n Embedding the Chunks. Turning chunks into vectors with Senten ce Transformers.\n Retrieval. Embedding the query and finding the most rel evant chunks via cosine similarity.\n Building the Prompt. Assembling labe lled context and instructing the model to cite section and page.\n Generat ing the Answer. Passing the retrieved context to an LLM for a grounded res ponse.\n Summary and Q&A. Key takeaways and open discussion.\n\n\n \n\nTh e project is co-funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect th ose of the European Union or the European Commission. Neither the European Union nor the granting authority can be held responsible for them. \nhttp s://events.grnet.gr/event/210/ LOCATION: URL:https://events.grnet.gr/event/210/ END:VEVENT BEGIN:VEVENT SUMMARY:PHAROS Training Series - Course 9 "RAG End-to-End: Architecture\, Retrieval\, Generation and Evaluation" DTSTART;VALUE=DATE-TIME:20260707T080000Z DTEND;VALUE=DATE-TIME:20260707T121500Z DTSTAMP;VALUE=DATE-TIME:20260716T014300Z UID:indico-event-213@events.grnet.gr DESCRIPTION:\n\nPHAROS AI Factory announces the 9th Course of its Training Series\, under the title "RAG End-to-End: Architecture\, Retrieval\, Gene ration and Evaluation"\, under the topic AI4LanguageCulture\, held online via Zoom.  \n\nDate: July 7th\, 2026\, at 11:00 EEST \n\nLocation: On line via Zoom\n\nPresentation Language: Greek\n\nAudience:  Machine Learn ing Engineers\, AI Engineers\, Data Scientists\, Academic Researchers\, La nguage and Culture Experts  \n\nPrerequisites: Basic Python knowledge \n \nLearning Objectives: \n\n\n Explain the core principles and architectur e of Retrieval-Augmented Generation systems.\n Understand why RAG improves factuality\, grounding\, transparency and access to external knowledge.\n Describe the main RAG pipeline stages\, from ingestion and preprocessing to retrieval and response generation.\n Identify design choices for chunki ng\, embeddings\, vector storage\, retrieval\, prompting and answer ground ing.\n Evaluate retrieval quality\, generation quality and end-to-end RAG behaviour.\n\n\nLearning Outcomes: \n\n\nAfter completing the course\, pa rticipants will have: \n\n\n\n\n \n A clear understanding of the main com ponents and design paradigms of RAG systems. \n \n\n\n\n\n\n \n Practical familiarity with document preparation\, chunking\, embedding generation\, vector indexing and similarity-based retrieval. \n \n\n\n\n\n\n \n Han ds-on experience in constructing a working RAG pipeline using Python and c ontemporary tools. \n \n\n\n\n\n\n \n The ability to connect retrieved ev idence with LLM-based answer generation in a grounded and transparent mann er. \n \n\n\n\n\n\n \n Familiarity with evaluation approaches for retriev al\, generation\, faithfulness\, groundedness and overall RAG performanc e. \n \n\n\n\n\n\n \n An understanding of how RAG can support Greek-langu age applications\, including public-service information retrieval and conv ersational assistance. \n \n\n\n\n\n\n \n The skills to analyse\, evalu ate and improve RAG systems for real-world deployment \n \n\n\n\nInstruct ors' profiles:\n\n\n George Drosatos\, ATHENA RC\n\n\nGeorge Drosatos is a Principal Researcher\, Researcher Grade B\, at the Institute for Languag e and Speech Processing of the Athena Research Center\, with expertise in privacy technologies\, information retrieval\, content analysis\, informat ion security and biomedical informatics. He holds a Diploma\, MSc and PhD in Electrical and Computer Engineering from Democritus University of Thrac e. He has participated in more than 20 national and European research proj ects and has extensive teaching experience in undergraduate and postgradua te courses at Greek and international universities. His research focuses o n privacy-enhancing technologies\, secure data analysis\, trustworthy AI a nd data-driven systems. He has authored more than 76 publications\, with o ver 1\,800 citations\, h-index 23 and i10-index 37. He has also served as Guest Editor in multiple Special Issues and as Secretary General of EAMBES from 2023 to 2025. \n\nMore information: https://www.drosatos.info.\n\n\n Sotiris Gyftopoulos\, ATHENA RC\n\n\nSotiris Gyftopoulos is a Scientific Associate at the Institute for Language and Speech Processing (ILSP) of t he Athena Research Center. He holds a degree in Computer Science from the University of Crete and a PhD from the Department of Electrical and Comput er Engineering at Democritus University of Thrace\, with his doctoral rese arch focusing on influence analysis in social networks. His expertise lies at the intersection of Natural Language Processing (NLP)\, statistical da ta analysis and social network modelling. With extensive experience in nat ional and European research projects\, Dr Gyftopoulos has also taught grad uate-level courses on data analysis and database systems. His scientific w ork has been published in international journals and conference proceeding s\, with emphasis on information diffusion and influence analysis through stochastic processes and advanced machine learning techniques.\n\n \n\nNo te: Please enter your institutional/corporate email when registering.\nhtt ps://events.grnet.gr/event/213/ LOCATION: URL:https://events.grnet.gr/event/213/ END:VEVENT BEGIN:VEVENT SUMMARY:Navigate through our learning platform and access opportunities DTSTART;VALUE=DATE-TIME:20260708T090000Z DTEND;VALUE=DATE-TIME:20260729T110000Z DTSTAMP;VALUE=DATE-TIME:20260716T014300Z UID:indico-event-214@events.grnet.gr DESCRIPTION:The Pharos Training Platform provides SMEs and start-ups with a structured\, self-paced learning journey designed to build practical exp ertise in AI and High-Performance Computing (HPC). Covering everything fro m machine learning fundamentals to large language models and sector-specif ic applications\, the platform combines curated learning pathways\, hands- on tutorials\, and live\, business-focused sessions.\n\nParticipants gain the knowledge and skills to develop\, deploy\, and scale AI solutions\, in cluding running AI workloads on European supercomputers\, with practical g uidance on accessing and making effective use of available computing resou rces.\n \n\nWhat the Pharos Training Platform includes\n\n\n AI and machi ne learning fundamentals – Build a strong foundation with courses coveri ng exploratory data analysis\, data preprocessing\, and supervised and uns upervised learning\, preparing your team for more advanced AI applications .\n Deep learning and Generative AI – Develop practical expertise in neu ral networks\, computer vision\, natural language processing (NLP)\, and l arge language models (LLMs)\, with hands-on training in model fine-tuning and Retrieval-Augmented Generation (RAG).\n Industry-focused AI applicatio ns – Explore real-world AI use cases across sectors including healthcare \, sustainability\, language\, and culture\, complemented by business case studies and expert talks demonstrating how organisations successfully dep loy AI in production environments.\n Access to European HPC resources – Learn how to scale AI workloads on High-Performance Computing (HPC) infras tructure and receive guidance on applying for compute resources through Eu roHPC AI Factory access programmes\, including Playground for rapid entry- level access and Fast Lane for medium-scale allocations designed for SMEs and start-ups.\n\n\nNeed personalised guidance?\n\nBook a free 30-minute o ne-to-one consultation\, available every Wednesday in July (8\, 15\, 22 an d 29 July)\, from 12:00 to 14:00\, to discuss your learning goals\, AI use cases\, or access to HPC resources.\n\nNote: Please enter your institutio nal/corporate email when registering.\nhttps://events.grnet.gr/event/214/ LOCATION: URL:https://events.grnet.gr/event/214/ END:VEVENT BEGIN:VEVENT SUMMARY:PHAROS Training Series - Course 10 "Time-Series Forecasting and Re newable Energy" DTSTART;VALUE=DATE-TIME:20260710T090000Z DTEND;VALUE=DATE-TIME:20260710T103000Z DTSTAMP;VALUE=DATE-TIME:20260716T014300Z UID:indico-event-215@events.grnet.gr DESCRIPTION:\n\nPHAROS AI Factory announces the 10th Course of its Trainin g Series\, under the title "Time-Series Forecasting and Renewable Energy"\ , under the topic Machine Learning\, held online via Zoom.\n\nDate: July 10th\, 2026\, at 12:00 EEST \n\nLocation: Online via Zoom\n\nPresentatio n Language: Greek\n\nAudience:  ML Engineers\, AI Engineers\, Data Scient ists\, Academic Researchers \n\nPrerequisites:  Machine learning\, Python \n\nLearning Objectives: \n\n- Develop and evaluate Machine Learning mo dels for day-ahead solar energy production forecasting. \n- Interpret for ecasting results using appropriate performance metrics and validation te chniques. \n- Understand the role of accurate renewable energy forecast ing in electricity markets and energy system operations. \n- Describe how renewable energy aggregators use forecasting to support market participat ion and grid stability. \n- Recognize the importance of effective data vi sualization and user interfaces for communicating forecasting results. \n - Apply best practices for designing end-to-end forecasting pipelines\, fr om data collection to model deployment. \n\nInstructors' profiles:\n\n\n I oannis Vlahavas\, AUTH\n\n\nIoannis Vlahavas is a professor at the Departm ent of Informatics at the Aristotle University of Thessaloniki. He receive d his Ph.D. degree in Logic Programming Systems from the same University i n 1988. He has been a visiting scholar at the Department of CS at Purdue U niversity and in 2017 elected EurAI Fellow from the European Association f or Artificial Intelligence. He specializes in knowledge based and AI (mach ine learning) systems and he has published more than 400 papers and book c hapters and co-authored 9 books in these areas. Google scholar gives a num ber exceeding 21600 citations and an h-index of 60. He has successfully su pervised 19 PhD students\, and he has been involved in more than 50 resear ch and development projects\, leading most of them. He was Chairman of the school of Informatics at the Aristotle University (2013-2017)\, member of the steering committee and Dean of the School of Science and Technology o f the International Hellenic University (2007-2016) and member of the Sect oral Scientific Council in Artificial Intelligence (2025-today). He is cur rently leading the Intelligent Systems Lab. \n\nDetails in https://intell igence.csd.auth.gr/people/vlahavas/\n\nLinkedin: https://www.linkedin.com/ in/ioannis-vlahavas/\n\n\n Vasileios Kochliaridis\, AUTH\n\n\nVasileios Ko chliaridis graduated with a degree in Computer Engineering and Information Technology from the University of Ioannina\, and since 2021 has been a me mber of the Intelligent Systems Lab (ISL) and a Ph.D. candidate in the fie ld of Artificial Intelligence under the supervision of Professor Vlachava. The focus of his doctoral research is the development of intelligent agen ts for autonomous systems\, with a particular emphasis on financial data\, as well as on autonomous and intelligent systems. During his doctoral stu dies\, he has gained both technological and academic experience through re search projects and publications. In the technological field\, he has succ essfully contributed in the role of data scientist and deep learning resea rcher. Since 2022\, he has been the lead researcher in the development of a tool for time series forecasting and the generation of synthetic data  for financial metrics\, as well as urban environment data and images\, usi ng GAN networks to train agents for the navigation of autonomous vehicles within a simulation.\n\nMore information on his page: https://intelligenc e.csd.auth.gr/people/vasileios-kochliaridis/\n\n\n Anestis Ampatzidis\, EY & AUTH\n\n\nAnestis is an AI Consultant at EY and an Artificial Intellige nce Researcher at the Intelligent Systems Laboratory. He holds a Bachelor' s degree in Computer Science and a Master's degree in Data and Web Science from the Aristotle University of Thessaloniki\, with excellent academic p erformance. His expertise focuses on the development of integrated Artific ial Intelligence systems for the energy sector. He has extensive research and professional experience in time series forecasting\, for both energy p roduction and demand\, utilizing advanced Machine Learning models. Additio nally\, he has a strong background as a Machine Learning Engineer\, focusi ng on transforming models into comprehensive end-to-end solutions. He desi gns scalable cloud infrastructures and ensures the seamless integration of algorithms into production systems\, delivering reliable real-time predic tions.\n\nLinkedIn: https://www.linkedin.com/in/anestis-ampatzidis-a044153 01/ \n\nNote: Please enter your institutional/corporate email when regist ering.\n\n \nhttps://events.grnet.gr/event/215/ LOCATION: URL:https://events.grnet.gr/event/215/ END:VEVENT BEGIN:VEVENT SUMMARY:HPC Training Series - Course 22 "HPC Fundamentals\, Slurm\, Open O nDemand & XDMoD" DTSTART;VALUE=DATE-TIME:20260713T090000Z DTEND;VALUE=DATE-TIME:20260713T114000Z DTSTAMP;VALUE=DATE-TIME:20260716T014300Z UID:indico-event-218@events.grnet.gr DESCRIPTION:\n\nEuroCC@Greece announces the 22nd Course of HPC Training Series with the subject "HPC Fundamentals\, Slurm\, Open OnDemand & XDMoD ".\n\nDate: July 13th\, 2026\, at 12:00 EEST \n\nLocation: Online via Zo om  \n\nPresentation Language: Greek\n\nAudience: This course is ideal for HPC Engineers and Academic Researchers.\n\nLearning Objectives:\n\nBy the end of the seminar\, participants will be able to:\n\n\n Explain the r ole of High Performance Computing in AI\, science\, and engineering worklo ads.\n Identify key HPC components\, including nodes\, CPUs\, GPUs\, memor y\, storage\, and networking.\n Understand parallel computing concepts suc h as tasks\, threads\, cores\, speedup\, and scaling.\n Use basic Slurm co ncepts to request resources\, submit jobs\, and monitor workloads.\n Acces s HPC resources through Open OnDemand for interactive sessions\, file mana gement\, and job submission.\n Recognize when to use CPUs\, GPUs\, OpenMP\ , MPI\, or CUDA for different computational tasks.\n\n\nPrerequisites: Ba sic programming knowledge (preferably in C\, C++\, or Fortran).\n\nInstruc tors’ Short Bios:\n\n\n Alexandra Charalambidou is an HPC engineer at th e Digital Governance Unit of Aristotle University of Thessaloniki. She sup ports users of the University's HPC infrastructure\, and moreover designs and maintains HPC related services. She has several years of experience in creating user workflows\, porting\, profiling\, benchmarking and optimisi ng scientific applications on HPC resources. \n Anastasia Makri is an HPC team member at the Digital Governance Unit of Aristotle University of Thes saloniki and senior student at the the Department of Informatics of Aristo tle University\, supporting users of the University's HPC infrastructure a nd contributing to the improvement of the overall user experience. Her wor k focuses on user support\, troubleshooting\, software deployment and main tenance\, as well as scripting for the automation of administrative and op erational workflows in HPC environments. \n Paschalis Korosoglou is an HPC engineer at the Digital Governance Unit of the Aristotle University of Th essaloniki. He has participated in major national and European projects\, such as EGI\, HP-SEE\, PRACE and HellasHPC\, mainly in user services provi sioning and application support activities. In this respect he holds a sol id background on scientific applications porting\, profiling\, optimisatio n and benchmarking activities and has developed under several occasions co de optimisation improvements related to hybrid parallel implementations (M PI/OpenMP) and parallel I/O best practices. \n Vassilis Asteriou is an HPC and DevOps engineer at the Digital Governance Unit of Aristotle Universit y of Thessaloniki and a PhD candidate in communications networks at the De partment of Informatics of Aristotle University. He received the B.S. degr ee in Informatics at the Aristotle University of Thessaloniki\, Greece\, i n 2020. His professional experience includes HPC and Cloud infrastructure and services deployment and management\, installation\, distribution and t roubleshooting of scientific software and HPC user support. He has partici pated in EuroCC 2.\n Dr. Nikos Bakas is a Senior Data Scientist at GRNET w ith a broad background in Artificial Intelligence. He has authored numerou s publications across AI thematic areas including Machine Learning\, Numer ical Methods\, Optimization\, and Large Language Models. He has served as principal investigator\, researcher\, and coordinator in multiple projects at research centers and universities. Dr. Bakas holds a Ph.D. from the Na tional Technical University of Athens and has long-standing teaching exper ience. He also brings extensive programming expertise in a wide range of l anguages and frameworks\, and the training seminars he has organized have reached a broad community of engineers.\n\n\n \n\nNote: Please enter your institutional/corporate email when registering.\n\n \n\n \n\n---\n\n \ n\n\n\nEuroCC 3 has received funding from the European High-Performance Co mputing Joint Undertaking (JU) under Grant Agreement No. 101306701. The JU receives support from the European Union‘s Digital Europe Programme and Germany\, Albania\, Austria\, Belgium\, Bosnia and Herzegovina\, Bulgaria \, Croatia\, Cyprus\, Czechia\, Denmark\, Estonia\, Finland\, France\, Gre ece\,Hungary\, Iceland\, Ireland\, Italy\, Latvia\, Lithuania\, Luxembourg \, Malta\, Montenegro\, the Netherlands\, North Macedonia\, Norway\, Polan d\, Portugal\, Romania\, Serbia\, Slovakia\, Slovenia\, Spain\, Sweden\, T ürkiye\, and Kosovo. \n\nFunded by the European Union. Views and opinion s expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or EuroHPC Joint Undertaking. Neither the European Union nor the EuroHPC Joint Undertaking can be held responsi ble for them.\n\n \nhttps://events.grnet.gr/event/218/ LOCATION: URL:https://events.grnet.gr/event/218/ END:VEVENT BEGIN:VEVENT SUMMARY:PHAROS Training Series - Course 11 "Assessing and Mitigating Priva cy Risks in Machine Learning and Data-Intensive Environments" DTSTART;VALUE=DATE-TIME:20260714T090000Z DTEND;VALUE=DATE-TIME:20260714T120000Z DTSTAMP;VALUE=DATE-TIME:20260716T014300Z UID:indico-event-217@events.grnet.gr DESCRIPTION:\n\nPHAROS AI Factory announces the 11th Course of its Trainin g Series\, under the title "Assessing and Mitigating Privacy Risks in Mach ine Learning and Data-Intensive Environments"\, under the topic AI Ethics \, held online via Zoom. \n\nDate: July 14th\, 2026\, at 12:00 EEST \n\n Location: Online via Zoom\n\nPresentation Language: Greek\n\nAudience: C ompliance Officers\, Healthcare Experts \n\nPrerequisites: AI Ethics \n \nLearning Objectives: \n\n• Understand the privacy challenges introduc ed by modern data sharing and machine learning applications. \n• Learn the legal and technical differences between anonymization and pseudonymiza tion under GDPR. \n• Become familiar with established anonymization tec hniques\, including k-anonymity and km-anonymity\, and understand their ap plicability and limitations. \n• Understand the principles of different ial privacy and its role in privacy-preserving data analysis and machine l earning. \n• Gain practical experience using the Amnesia anonymization platform. \n• Understand the most important privacy attacks targeting m achine learning models and training datasets. \n• Learn about privacy-p reserving machine learning approaches\, including federated learning and d ifferential privacy.  \n\nInstructor's profile:\n\n\n Manolis Terrovitis \, Athena RC\n\n\nDr. Manolis Terrovitis is a Research Director at the Inf ormation Management Systems Institute (IMSI) of the Athena Research Center in Athens\, Greece. He has held several leadership positions\, including serving as President of the Hellenic Accreditation System and as a member of the Board of Directors of Information Society S.A.\, the Greek public o rganization responsible for the procurement of ICT services and products f or the public sector. \nHe has more than 20 years of experience in data m anagement research and in the design and development of large-scale data m anagement systems. His primary research interests include privacy-preservi ng data publishing\, data anonymization\, and big data management. His wor k has been published in leading data management venues\, including PVLDB\, VLDB Journal\, IEEE ICDE\, IEEE TKDE\, and others. \nManolis has extensi vely worked on data anonymization algorithms and leads the development of Amnesia (https://amnesia.openaire.eu)\, an open-source data anonymization platform widely used by researchers and organizations for privacy-preservi ng data sharing.\n\n \n\nNote: Please enter your institutional/corporate email when registering.\nhttps://events.grnet.gr/event/217/ LOCATION: URL:https://events.grnet.gr/event/217/ END:VEVENT BEGIN:VEVENT SUMMARY:PHAROS Training Series - Course 12 "Compute-Efficient Methods for Large Language Models" DTSTART;VALUE=DATE-TIME:20260717T080000Z DTEND;VALUE=DATE-TIME:20260717T110000Z DTSTAMP;VALUE=DATE-TIME:20260716T014300Z UID:indico-event-216@events.grnet.gr DESCRIPTION:\n\nPHAROS AI Factory announces the 12th Course of its Trainin g Series\, under the title "Compute-Efficient Methods for Large Language M odels"\, under the topic LLMs\, organised in collaboration with Pharos-CY \, held online via Zoom.  \n\nDate: July 17th\, 2026\, at 11:00 EEST \n \nLocation: Online via Zoom\n\nPresentation Language: English\n\nAudience : Data Scientists\, ML Engineers\, AI Engineers\, Academic Researchers\n\ nLevel: Intermediate\n\nPrerequisites: Machine Learning\, Python\n\nLearn ing Objectives: \n\nBy the end of the seminar\, participants will be able to:\n\n\n Identify key efficiency methods for training\, fine-tuning\, an d inference.\n Describe how LoRA enables parameter-efficient fine-tuning o f LLMs.\n Apply a basic Hugging Face workflow for dataset preparation\, tr aining\, evaluation\, and inference.\n Compare trade-offs between model pe rformance\, cost\, memory use\, and deployment efficiency.\n\n\nInstructor s’ Short Bios:\n\n\n Professor Constantine Dovrolis is Director of the C omputation-based Science and Technology Research Center (CaSToRC) at The C yprus Institute and\, starting in September 2026\, XM Chair in Artificial Intelligence at the University of Cyprus. He served on the faculty of the School of Computer Science at the Georgia Institute of Technology from 200 2 to 2025 and is an ACM Distinguished Member. His research spans machine l earning\, network science\, and data-driven modeling\, with a recent focus on neuro-inspired artificial intelligence. A central question in his work is how principles underlying the structure and function of brain networks —such as sparsity\, modularity\, plasticity\, and hierarchy—can guide the design of more adaptive\, efficient\, and interpretable learning syste ms. His work has appeared at venues including ICML\, NeurIPS\, CVPR\, and TMLR. He also collaborates broadly across neuroscience\, biology\, medicin e\, and climate science\, with funding from NSF\, NIH\, DOE\, DARPA\, Hori zon Europe\, and Cyprus’ RIF.\n\n\n \n\n\n Dr. Nikos Bakas is a Senior Data Scientist at GRNET with a broad background in Artificial Intelligence . He has authored numerous publications across AI thematic areas including Machine Learning\, Numerical Methods\, Optimization\, and Large Language Models. He has served as principal investigator\, researcher\, and coordin ator in multiple projects at research centers and universities. Dr. Bakas holds a Ph.D. from the National Technical University of Athens and has lon g-standing teaching experience. He also brings extensive programming exper tise in a wide range of languages and frameworks\, and the training semina rs he has organized have reached a broad community of engineers.\n\n\n \n \n\n Roman Dolgopolyi is an Artificial Intelligence Developer at GRNET. He has extensive experience in training\, fine-tuning\, and benchmarking bot h Large Language Models and Vision-Language Models. His previous work has been recognized in reputable academic journals and venues\, including Spri nger Nature publications and the EMCIS Conference.\n\n\n \n\nNote: Please enter your institutional/corporate email when registering.\n\n \nhttps:/ /events.grnet.gr/event/216/ LOCATION: URL:https://events.grnet.gr/event/216/ END:VEVENT BEGIN:VEVENT SUMMARY:BioExcel / EuroCC workshop on Biomolecular Simulations: Aegean Edi tion DTSTART;VALUE=DATE-TIME:20261118T074500Z DTEND;VALUE=DATE-TIME:20261120T124500Z DTSTAMP;VALUE=DATE-TIME:20260716T014300Z UID:indico-event-211@events.grnet.gr DESCRIPTION:\n\nBioExcel / EuroCC workshop on Biomolecular Simulations: Ae gean Edition \n\nDate: November 18-20\, 2026\n\nLocation: Biomedical Rese arch Foundation of the Academy of Athens (BRFAA)\n\nType of the event:  H ybrid (30 on-site + 10 online participants). Remote participants from the Aegean region and other areas outside Athens will be given priority.\n\nP resentation Language: English\n\nAudience: Suitable for  graduate student s\, postdocs\, PIs.\n\nParticipation: Open application with a selection. 3 0 applicants will be selected for on-site participation and 10 on-line. Ap plicants from Greece and Türkiye will be prioritised for on-site particip ation. We can only accept participants from EuroHPC Joint Undertaking mem ber institutions. \n\nWorkshop Description: This three-day workshop cover s biomolecular modeling\, simulation workflows and integrative structural biology. The event pairs lectures with hands-on tutorials using Jupyter No tebooks and web portals across three methods: GROMACS\, BioBB and HADDOCK . \n\n \n\nAgenda breakdown:\n\nModule 1: Introduction to MD & Trajectory Analysis (GROMACS)\n\n\n \n Instructors: Alessandra Villa (KCSC)\n \n \n Date: 18 November (Lecture & Tutorial)\n \n\n\nThis module covers classica l force fields\, boundary conditions and numerical integration used to sim ulate macromolecular trajectories.\n\nParticipants will learn to:\n\n\n \n Prepare and check an experimental PDB structure.\n \n \n Define topologie s\, parameterize molecules\, and select force fields.\n \n \n Solvate the system and neutralize charges with ions.\n \n \n Run energy minimization\, along with NVT and NPT equilibration phases.\n \n \n Run production MD an d analyze trajectories via Root Mean Square Deviation (RMSD)\, Root Mean S quare Fluctuation (RMSF) and Radius of Gyration (Rg​).\n \n\n\nModule 2: Automated Biomolecular Workflows (BioBB)\n\n\n \n Instructor: Adam Hospit al (IRB Barcelona)\n \n \n Date: 19 November (Lecture & Tutorial)\n \n\n\n This module introduces the BioExcel Building Blocks (BioBB) Python library to build automated\, reproducible simulation pipelines that follow FAIR d ata principles.\n\nParticipants will learn to:\n\n\n \n Use BioBB modules inside Jupyter Notebooks to link separate simulation steps.\n \n \n Automa te structure checking\, system preparation\, solvation\, equilibration\, a nd GROMACS execution blocks.\n \n \n Handle parameter adjustments programm atically\, track data provenance\, and scale workflows across computing se tups.\n \n\n\nModule 3: Integrative Structural Biology: MD & HADDOCK & Cry o-EM\n\n\n \n Instructors: Alexandre Bonvin (Utrecht University)\, Panagio tis Kastritis (MLU Halle-Wittenberg)\n \n \n Dates: 19–20 November (Lect ures & Tutorials)\n \n\n\nThis module covers information-driven docking to model large macromolecular complexes by combining electron microscopy dat a with computational prediction in HADDOCK.\n\nParticipants will learn to: \n\n\n \n Preprocess Data: Crop Cryo-EM density maps and use PowerFit 6D c ross-correlation searches to locate subunit centroids.\n \n \n Restraint-D riven Docking: Convert centroid coordinates into distance restraints in HA DDOCK to drive initial rigid-body docking.\n \n \n Refinement & Scoring: R un semi-flexible simulated annealing in explicit solvent\, apply density s coring terms\, and cluster the resulting structural configurations.\n \n\n \n \n\nImportant dates:\n\n\n 10 July 2026: Registration opens\n 25 Septe mber 2026: Registration closes\n 5 October 2026: Notification of selected participants\n 10 October 2026: Deadline for confirmation of attendance \n \n\nFor any questions regarding the workshop or the application / registra tion process\, please contact us at: zoe@bioacademy.gr \n\n \n\n---\n\n  \n\n \n\n\n\nEuroCC 3 has received funding from the European High-Perfo rmance Computing Joint Undertaking (JU) under Grant Agreement No. 10130670 1. The JU receives support from the European Union‘s Digital Europe Prog ramme and Germany\, Albania\, Austria\, Belgium\, Bosnia and Herzegovina\, Bulgaria\, Croatia\, Cyprus\, Czechia\, Denmark\, Estonia\, Finland\, Fra nce\, Greece\,Hungary\, Iceland\, Ireland\, Italy\, Latvia\, Lithuania\, L uxembourg\, Malta\, Montenegro\, the Netherlands\, North Macedonia\, Norwa y\, Poland\, Portugal\, Romania\, Serbia\, Slovakia\, Slovenia\, Spain\, S weden\, Türkiye\, and Kosovo. \n\nFunded by the European Union. Views an d opinions expressed are however those of the author(s) only and do not ne cessarily reflect those of the European Union or EuroHPC Joint Undertaking . Neither the European Union nor the EuroHPC Joint Undertaking can be held responsible for them. \nhttps://events.grnet.gr/event/211/ LOCATION: URL:https://events.grnet.gr/event/211/ END:VEVENT END:VCALENDAR