English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Teaching is critical to human society: it is with teaching that prospectivestudents are educated and human civilization can be inherited and advanced. Agood teacher not only provides his/her students with qualified teachingmaterials (e.g., textbooks), but also sets up appropriate learning objectives(e.g., course projects and exams) considering different situations of astudent. When it comes to artificial intelligence, treating machine learningmodels as students, the loss functions that are optimized act as perfectcounterparts of the learning objective set by the teacher. In this work, weexplore the possibility of imitating human teaching behaviors by dynamicallyand automatically outputting appropriate loss functions to train machinelearning models. Different from typical learning settings in which the lossfunction of a machine learning model is predefined and fixed, in our framework,the loss function of a machine learning model (we call it student) is definedby another machine learning model (we call it teacher). The ultimate goal ofteacher model is cultivating the student to have better performance measured ondevelopment dataset. Towards that end, similar to human teaching, the teacher,a parametric model, dynamically outputs different loss functions that will beused and optimized by its student model at different training stages. Wedevelop an efficient learning method for the teacher model that makes gradientbased optimization possible, exempt of the ineffective solutions such as policyoptimization. We name our method as "learning to teach with dynamic lossfunctions" (L2T-DLF for short). Extensive experiments on real world tasksincluding image classification and neural machine translation demonstrate thatour method significantly improves the quality of various student models.

Learning to Teach with Dynamic Loss Functions