Aravind Putrevu
Devrel
ค้นพบ, เชื่อมต่อ, เติบโต
พร้อมที่จะเริ่มต้นใช้งาน AI แล้วหรือยัง Microsoft Reactor มีกิจกรรม การฝึกอบรม และทรัพยากรของชุมชนเพื่อช่วยให้บริษัทสตาร์ทอัพ ผู้ประกอบการ และนักพัฒนาสร้างธุรกิจใหม่บนเทคโนโลยี AI เข้าร่วมกับเรา!
ค้นพบ, เชื่อมต่อ, เติบโต
พร้อมที่จะเริ่มต้นใช้งาน AI แล้วหรือยัง Microsoft Reactor มีกิจกรรม การฝึกอบรม และทรัพยากรของชุมชนเพื่อช่วยให้บริษัทสตาร์ทอัพ ผู้ประกอบการ และนักพัฒนาสร้างธุรกิจใหม่บนเทคโนโลยี AI เข้าร่วมกับเรา!
26 กุมภาพันธ์, 2567 | 12:00 หลังเที่ยง - 1:00 หลังเที่ยง (UTC) เวลามาตรฐานสากล
หัว ข้อ: Data Science & Machine Learning
ภาษา: ภาษาอังกฤษ
Labeled data powers AI/ML in the enterprise, but real-world datasets have been found to contain between 7-50% annotation errors. Imperfectly labelled text data hampers ML models' training (and evaluation) across tasks like intent recognition, entity recognition, and sequence generation. Although pretrained LLMs are equipped with a lot of world knowledge, their performance is adversely affected by noisy training data (as noted by OpenAI).
In this talk, we illustrate data-centric techniques to mitigate the effect of label noise without changing any code related to model architecture, hyperparameters, or training. These data quality improvement techniques should thus remain applicable even for future advanced LLMs like GPT-10.
ลำโพง
สําหรับคําถามโปรดติดต่อเราที่ reactor@microsoft.com