BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:America/Los_Angeles BEGIN:DAYLIGHT DTSTART:20260308T030000 TZOFFSETFROM:-0800 TZOFFSETTO:-0700 RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3 TZNAME:PDT END:DAYLIGHT BEGIN:STANDARD DTSTART:20261101T010000 TZOFFSETFROM:-0700 TZOFFSETTO:-0800 RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11 TZNAME:PST END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20260606T220324Z UID:EC2A74AF-6BC4-4A95-A9A4-E67445EBAF13 DTSTART;TZID=America/Los_Angeles:20260604T180000 DTEND;TZID=America/Los_Angeles:20260604T193000 DESCRIPTION:Abstract: Reinforcement learning (RL) is usually introduced thr ough games and simulations\, but many real-world problems also involve seq uential decisions under uncertainty. In this talk\, I will present three a pplications of RL in practical machine learning systems. First\, I will di scuss how RL can be used for time-series anomaly detection\, where an agen t learns to balance missed anomalies and false alarms over time. Second\, I will describe multimodal RL in the presence of adversarial noise\, and h ow robustness issues arise when combining signals from different data sour ces. Finally\, I will show how (deep) RL can be applied to group recommend ation systems\, where the goal is to optimize long-term engagement while a ccounting for diverse user preferences within a group. Throughout the talk \, the focus will be on the high-level ideas\, design choices\, and lesson s learned\, rather than algorithmic details. I will highlight common chall enges across these projects—such as reward design\, stability\, and eval uation—and discuss open questions for deploying RL in real-world setting s.\n\nSpeaker(s): Banafsheh\, \n\nVirtual: https://events.vtools.ieee.org/ m/561747 LOCATION:Virtual: https://events.vtools.ieee.org/m/561747 ORGANIZER:edward.c.epp@ieee.org SEQUENCE:83 SUMMARY:Reinforcement learning (RL) URL;VALUE=URI:https://events.vtools.ieee.org/m/561747 X-ALT-DESC:Description: <br /><p>Abstract: Reinforcement learning (RL) is u sually introduced through games and simulations\, but many real-world prob lems also involve sequential decisions under uncertainty. In this talk\, I will present three applications of RL in practical machine learning syste ms. First\, I will discuss how RL can be used for time-series anomaly dete ction\, where an agent learns to balance missed anomalies and false alarms over time. Second\, I will describe multimodal RL in the presence of adve rsarial noise\, and how robustness issues arise when combining signals fro m different data sources. Finally\, I will show how (deep) RL can be appli ed to group recommendation systems\, where the goal is to optimize long-te rm engagement while accounting for diverse user preferences within a group . Throughout the talk\, the focus will be on the high-level ideas\, design choices\, and lessons learned\, rather than algorithmic details. I will h ighlight common challenges across these projects&mdash\;such as reward des ign\, stability\, and evaluation&mdash\;and discuss open questions for dep loying RL in real-world settings.</p>\n<p>&nbsp\;</p> END:VEVENT END:VCALENDAR