BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:America/New_York BEGIN:DAYLIGHT DTSTART:20260308T030000 TZOFFSETFROM:-0500 TZOFFSETTO:-0400 RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3 TZNAME:EDT END:DAYLIGHT BEGIN:STANDARD DTSTART:20261101T010000 TZOFFSETFROM:-0400 TZOFFSETTO:-0500 RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11 TZNAME:EST END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20260427T132154Z UID:96AF2956-61C5-49F3-936E-ED600B8B0B41 DTSTART;TZID=America/New_York:20260511T190000 DTEND;TZID=America/New_York:20260511T200000 DESCRIPTION:Cyberattacks are rapidly evolving in complexity\, scale\, and u npredictability\, posing serious threats to critical infrastructure system s. As organizations face increasingly sophisticated\, multi-vector attacks \, traditional security approaches are no longer sufficient. This highligh ts the growing importance of Cyber Threat Intelligence (CTI) in enabling p roactive and adaptive defense strategies.\n\nIn this talk\, I present an a dvanced deep learning–based intrusion detection framework that combines Convolutional Neural Networks (CNN) and Bidirectional Long Short-Term Memo ry (Bi-LSTM) models. This hybrid architecture is designed to effectively c apture both spatial and temporal patterns in dynamic network traffic\, imp roving the detection of complex cyber threats.\n\nThe model is trained and evaluated using the UNSW-NB15 dataset\, incorporating comprehensive prepr ocessing techniques such as data cleaning\, normalization\, and feature op timization. The proposed approach demonstrates exceptional performance\, a chieving 99.78% accuracy\, 98.67% precision\, 99.99% recall\, and a 98.89% F1-score significantly outperforming traditional machine learning and dee p learning models.\n\nThis session will provide insights into how AI-drive n cybersecurity solutions can enhance threat detection\, reduce false posi tives\, and strengthen resilience in critical infrastructure environments. Attendees will gain practical knowledge on applying deep learning techniq ues to modern cybersecurity challenges.\n\nCo-sponsored by: Future Network s AI/ML Working Group\n\nSpeaker(s): \, Siva\n\nVirtual: https://events.vt ools.ieee.org/m/551403 LOCATION:Virtual: https://events.vtools.ieee.org/m/551403 ORGANIZER:anushanerella@ieee.org SEQUENCE:43 SUMMARY:WIE Speaker Series 2026: AI and Cloud-Powered Security for Critical Infrastructure: Next-Generation Threat Detection and Defense URL;VALUE=URI:https://events.vtools.ieee.org/m/551403 X-ALT-DESC:Description: <br /><p>Cyberattacks are rapidly evolving in compl exity\, scale\, and unpredictability\, posing serious threats to critical infrastructure systems. As organizations face increasingly sophisticated\, multi-vector attacks\, traditional security approaches are no longer suff icient. This highlights the growing importance of Cyber Threat Intelligenc e (CTI) in enabling proactive and adaptive defense strategies.</p>\n<p>In this talk\, I present an advanced deep learning&ndash\;based intrusion det ection framework that combines Convolutional Neural Networks (CNN) and Bid irectional Long Short-Term Memory (Bi-LSTM) models. This hybrid architectu re is designed to effectively capture both spatial and temporal patterns i n dynamic network traffic\, improving the detection of complex cyber threa ts.</p>\n<p>The model is trained and evaluated using the UNSW-NB15 dataset \, incorporating comprehensive preprocessing techniques such as data clean ing\, normalization\, and feature optimization. The proposed approach demo nstrates exceptional performance\, achieving 99.78% accuracy\, 98.67% prec ision\, 99.99% recall\, and a 98.89% F1-score significantly outperforming traditional machine learning and deep learning models.</p>\n<p>This sessio n will provide insights into how AI-driven cybersecurity solutions can enh ance threat detection\, reduce false positives\, and strengthen resilience in critical infrastructure environments. Attendees will gain practical kn owledge on applying deep learning techniques to modern cybersecurity chall enges.</p> END:VEVENT END:VCALENDAR