Uniavisen
Københavns Universitet
Uafhængig af ledelsen

Foredrag

From Machine Learning to Machine Teaching

Foredrag — Hver anden tirsdag til frokost giver en DIKU forsker en præsentation af et specifikt forskningsområde. Foredragene vil give dig et indblik i DIKUs forskellige forskningsområder, motivere dig til at følge din interesse inden for datalogi og forberede dig til at vælge et emne til dit bachelorprojekt.

Info

Date & Time:

Place:
Lille UP1, DIKU, Universitetsparken 1, 2100 København Ø

Hosted by:
DIKU, Københavns Universitet

Cost:
Free

On Tuesday 25th of March from 12.15-13.00 in Small UP1, Bob Pepin from the ML section will speak at DIKU Bits.

Machine Teaching explores the opportunities offered by learning from carefully chosen examples provided by a teacher model, as opposed to examples sampled at random. Bob will introduce the basic ideas with algorithms for teaching and learning hyperplanes in arbitrary dimension and multivariate Gaussian variables. He will explain how Machine Teaching relates to popular learning methods in LLMs such as in-context learning and test-time compute and finish with open theoretical questions.

DIKU Bits is the perfect place to spark your inspiration, fuel interests, and get closer to research at DIKU. Whether you are considering a project out of scope, want to get inspired for an upcoming bachelor project or thesis, or simply find the topic exciting, this is the perfect opportunity for you to expand your horizons.

Seneste