Aravind Putrevu
Devrel
LEARN, CONNECT, BUILD
Klaar om aan de slag te gaan met AI en de nieuwste technologieën? Microsoft Reactor biedt evenementen, training en communitybronnen om ontwikkelaars, ondernemers en startups te helpen bouwen op AI-technologie en meer. Kom kijken.
LEARN, CONNECT, BUILD
Klaar om aan de slag te gaan met AI en de nieuwste technologieën? Microsoft Reactor biedt evenementen, training en communitybronnen om ontwikkelaars, ondernemers en startups te helpen bouwen op AI-technologie en meer. Kom kijken.
26 februari, 2024 | 12:00 p.m. - 1:00 p.m. (UTC) Coordinated Universal Time
Onderwerp: Datawetenschap & Machine Learning
Taal: Engels
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.
Sprekers
De onderstaande gebeurtenissen zijn ook interessant voor u. Zorg ervoor dat u onze Reactor-startpagina om alle beschikbare gebeurtenissen weer te geven.
Notatie:
Livestream
Onderwerp: Datawetenschap & Machine Learning
Taal: Engels
Notatie:
Livestream
Onderwerp: Datawetenschap & Machine Learning
Taal: Engels