BEGIN:VCALENDAR VERSION:2.0 PRODID:-//jEvents 2.0 for Joomla//EN CALSCALE:GREGORIAN METHOD:PUBLISH BEGIN:VTIMEZONE TZID:America/New_York BEGIN:STANDARD DTSTART:20241103T010000 RDATE:20250309T030000 TZOFFSETFROM:-0400 TZOFFSETTO:-0500 TZNAME:America/New_York EST END:STANDARD BEGIN:STANDARD DTSTART:20251102T010000 RDATE:20260308T030000 TZOFFSETFROM:-0400 TZOFFSETTO:-0500 TZNAME:America/New_York EST END:STANDARD BEGIN:STANDARD DTSTART:20261101T010000 RDATE:20270314T030000 TZOFFSETFROM:-0400 TZOFFSETTO:-0500 TZNAME:America/New_York EST END:STANDARD BEGIN:STANDARD DTSTART:20271107T010000 RDATE:20280312T030000 TZOFFSETFROM:-0400 TZOFFSETTO:-0500 TZNAME:America/New_York EST END:STANDARD BEGIN:STANDARD DTSTART:20281105T010000 RDATE:20290311T030000 TZOFFSETFROM:-0400 TZOFFSETTO:-0500 TZNAME:America/New_York EST END:STANDARD BEGIN:DAYLIGHT DTSTART:20240430T153000 RDATE:20241103T010000 TZOFFSETFROM:-0500 TZOFFSETTO:-0400 TZNAME:America/New_York EDT END:DAYLIGHT BEGIN:DAYLIGHT DTSTART:20250309T030000 RDATE:20251102T010000 TZOFFSETFROM:-0500 TZOFFSETTO:-0400 TZNAME:America/New_York EDT END:DAYLIGHT BEGIN:DAYLIGHT DTSTART:20260308T030000 RDATE:20261101T010000 TZOFFSETFROM:-0500 TZOFFSETTO:-0400 TZNAME:America/New_York EDT END:DAYLIGHT BEGIN:DAYLIGHT DTSTART:20270314T030000 RDATE:20271107T010000 TZOFFSETFROM:-0500 TZOFFSETTO:-0400 TZNAME:America/New_York EDT END:DAYLIGHT BEGIN:DAYLIGHT DTSTART:20280312T030000 RDATE:20281105T010000 TZOFFSETFROM:-0500 TZOFFSETTO:-0400 TZNAME:America/New_York EDT END:DAYLIGHT END:VTIMEZONE BEGIN:VEVENT UID:6c805fcd06f52f674e20595f9d0ee08b CATEGORIES:Special Seminar CREATED:20250428T091423 SUMMARY:Rates of convergence in Pool-Based Batch Active Learning LOCATION:Hill 423 DESCRIPTION:We consider a supervised machine learning scenario called batch active lear ning, where a learning agent adaptively issues batches of points to a label ing oracle. Sampling labels in batches is highly desirable in practice due to the smaller number of interactive rounds with the labeling oracle (often human beings or highly expensive pre-trained models). Yet, batch active le arning typically pays the price of reduced adaptivity, leading to suboptima l convergence results. We describe a solution which requires a careful trad e off between the informativeness of the queried points and their diversity . We theoretically investigate batch active learning in the practically rel evant scenario where the unlabeled pool of data is available beforehand (po ol-based batch active learning). We analyze a stage-wise greedy algorithm a nd show that, as a function of the label complexity, the excess risk of thi s algorithm matches the known minimax rates in a standard statistical learn ing setting with linear function spaces. Our results also exhibit a mild de pendence on the batch size. These results are then extended to hold for gen eral function spaces with similar algorithmics, but more involved analytica l tools. Joint with: Z. Wang and T. Zhang. X-ALT-DESC;FMTTYPE=text/html:
We consider a supervised machine learning scenario called batch active learning, where a learning agent adaptively is sues batches of points to a labeling oracle. Sampling labels in batches is highly desirable in practice due to the smaller number of interactive round s with the labeling oracle (often human beings or highly expensive pre-trai ned models). Yet, batch active learning typically pays the price of reduced adaptivity, leading to suboptimal convergence results. We describe a solut ion which requires a careful trade off between the informativeness of the q ueried points and their diversity. We theoretically investigate batch activ e learning in the practically relevant scenario where the unlabeled pool of data is available beforehand (pool-based batch active learning). We analyz e a stage-wise greedy algorithm and show that, as a function of the label c omplexity, the excess risk of this algorithm matches the known minimax rate s in a standard statistical learning setting with linear function spaces. O ur results also exhibit a mild dependence on the batch size. These results are then extended to hold for general function spaces with similar algorith mics, but more involved analytical tools. 
Joint with: Z. Wang and T. Zhang.
CONTACT:Claudio Gentile (Google Research) DTSTAMP:20260829T233623 DTSTART;TZID=America/New_York:20250501T153000 DTEND;TZID=America/New_York:20250501T163000 SEQUENCE:0 TRANSP:OPAQUE END:VEVENT END:VCALENDAR