From Weakly Supervised Learning to Active Labeling PhD Thesis - VIVIEN. CABANNES
-20% with code BOOKS
Shipping in 10-16 days
30-day return policy
Applied maths and machine computations have raised a lot of hope since the recent success of supervised learning. Many practitioners in industries have been trying to switch from their old paradigms to machine learning. Interestingly, those data scientists spend more time scrapping, annotating and cleaning data than fine-tuning models. This PhD thesis is motivated by the following question: can we derive a ... Full description
You May Also Like
Spanish Ab Initio for the IB Diploma Grammar and Skills Workbook Hodder Education Group
-20% with code BOOKS
In stock at our supplier
Five-Minute Activities A Resource Book of Short Activities
-20% with code BOOKS
In stock at our supplier
French B for the IB Diploma Second Edition Hodder Education Group
Laetitia Chanéac-Knight, Lauren Léchelle
-20% with code BOOKS
In stock at our supplier
Languages and Learners Making the Match: World Language Instruction in K-8 Classrooms and Beyond
Helena Anderson Curtain, Carol Ann Dahlberg
-20% with code BOOKS
In stock at our supplier
Teaching Languages to Young Learners
-20% with code BOOKS
In stock at our supplier
Government Communication Cases and Challenges
Karen Sanders, Maria Jose Canel
-20% with code BOOKS
In stock at our supplier
200 Quotes that Will Change the Way You Think
-20% with code BOOKS
In stock at our supplier
Make a Difference-- with the Power of Compassion
-20% with code BOOKS
In stock at our supplier
A Short Guide to Writing about Music
-20% with code BOOKS
In stock at our supplier
Script, Print and Letterforms in Global Contexts The Visual and the Material
Caroline Archer-Parré, Malcolm Dick, Sahar Afshar, Hazel Wilkinson, Wei Jin Darryl Lim, Vaibhav Singh
-20% with code BOOKS
In stock at our supplier
The Dewey-Soka Heritage and the Future of Education
Jason Goulah, Awad Ibrahim, Gonzalo Obelleiro, James W. Garrison, Jim Garrison
-20% with code BOOKS
In stock at our supplier
The Balkan Linguistic Area An Atlas of Variation and Contact
Evangelia Adamou, Andrey N. Sobolev
-20% with code BOOKS
In stock at our supplier
Engaging with Ethics in Multilingual Learning Communities
Theresa Austin, Hatice Çelebi
-20% with code BOOKS
In stock at our supplier
Gender-Inclusive Language Findings from 14 Languages and Open Research Questions
-20% with code BOOKS
In stock at our supplier
Total Participation Techniques Making Every Student an Active Learner
Pérsida Himmele, William Himmele
-20% with code BOOKS
In stock at our supplier
Gesture in Second Language Acquisition and Pedagogy
Laura M. Morett, Solène Inceoglu
-20% with code BOOKS
In stock at our supplier
China's Publishing Industry From Mao to the Market
-20% with code BOOKS
In stock at our supplier
Disaster Planning for Special Libraries
-20% with code BOOKS
In stock at our supplier
Concise History of the Language Sciences: From the Sumerians to the Cognitivists
-20% with code BOOKS
In stock at our supplier
Schaum's Easy Outline of German
-20% with code BOOKS
In stock at our supplier
Description
This question is approached through the lens of weakly supervised learning, assuming that the bottleneck of data collection lies in annotation. We model weak supervision as giving, rather than a unique target, a set of target candidates. We argue that one should look for an "optimistic" function that matches most of the observations. This allows us to derive a principle to disambiguate partial labels. We also discuss the advantage to incorporate unsupervised learning techniques into our framework, in particular manifold regularization approached through diffusion techniques, for which we derived a new algorithm that scales better with input dimension then the baseline method.
Finally, we switch from passive to active weakly supervised learning, introducing the "active labeling" framework, in which a practitioner can query weak information about chosen data. Among others, we leverage the fact that one does not need full information to access stochastic gradients and perform stochastic gradient descent.
More Information
| Author | VIVIEN. CABANNES |
|---|---|
| Publisher | Amazon Digital Services LLC - Kdp |
| Release year | 2022 |
| Cover type | Softcover |
| EAN | 9798831483567 |