learning loss for active learning
play

Learning Loss for Active Learning Rymarczyk D., Zieliski B., Tabor - PowerPoint PPT Presentation

Apr 05, 2024 •366 likes •802 views

Learning Loss for Active Learning Rymarczyk D., Zieliski B., Tabor J., Sadowski M., Titov M. Agenda 1. Active Learning introduction 2. Base methods in AL for Deep Learning 3. Learning Loss for Active Learning 4. Our ideas for Active

  1. Learning Loss for Active Learning Rymarczyk D., Zieliński B., Tabor J., Sadowski M., Titov M.

  2. Agenda 1. Active Learning introduction 2. Base methods in AL for Deep Learning 3. Learning Loss for Active Learning 4. Our ideas for Active Learning 5. Future plans 6. Bibliography

  3. Active Learning Labeled Unlabeled dataset

  4. Active Learning Labeled Unlabeled Label dataset

  5. Active Learning Labeled Unlabeled dataset Predicted Label

  6. Active Learning Labeled Unlabeled dataset Predicted Label True Label

  7. Active Learning Labeled Unlabeled Label dataset

  8. Challenges in Active Learning 1. Criteria on which the sample will be chosen to the labelling process. 2. How many samples should be included in the labelling process? 3. Is the oracle infallible? 4. Multi oracle scenarios. 5. Online learning. 6. Can we use unlabeled data? and how? 7. How does the oracle reckon the AL system? 8. ...

  9. Challenges in Active Learning 1. Criteria on which the sample will be chosen to the labelling process. 2. How many samples should be included in the labelling process? 3. Is the oracle infallible? 4. Multi oracle scenarios. 5. Online learning. 6. Can we use unlabeled data? and how? 7. How does the oracle reckon the AL system? 8. ...

  10. Base methods in AL for Deep Learning Random sampling to label Labeled Unlabeled dataset

  11. Base methods in AL for Deep Learning Core-Set approach - K-Greedy algorithm / KMeans++ Labeled Unlabeled dataset

  12. Base methods in AL for Deep Learning Core-Set approach - K-Greedy algorithm / KMeans++ Labeled Unlabeled dataset

  13. Base methods in AL for Deep Learning Uncertainty based approach - entropy 0.1 Labeled 0.4 Unlabeled dataset 0.3 0.9 0.1 0.2 0.4 0.5

  14. Base methods in AL for Deep Learning Uncertainty based approach - entropy 0.1 Labeled 0.4 Unlabeled dataset 0.3 0.9 0.1 0.2 0.4 0.5

  15. Base methods in AL for Deep Learning Experiment - learning episode: Labeled 1000 Unlabeled dataset (CIFAR10) Predicted Label To be Labeled True Label 1000

  16. Base methods in AL for Deep Learning Experiment - 10 x learning episodes: Labeled 10000 Unlabeled dataset (CIFAR10) Predicted Label

  17. Base methods in AL for Deep Learning Experiment - Results

  18. Learning Loss for Active Learning Motivation: 1. None of the basic methods use information from inner layers of NN. 2. Best measure of NN error is value of loss function . 3. More advanced methods requires: a. modifications of the architecture, b. training another neural network , c. training generative model, d. finding adversarial examples, e. bayesian deep learning, f. model ensembles .

  19. Learning Loss for Active Learning Architecture modifications - learning loss module:

  20. Learning Loss for Active Learning Architecture modifications - learning loss module:

  21. Learning Loss for Active Learning Loss function for learning the loss

  22. Learning Loss for Active Learning Results from the paper: Experiment details on CIFAR10: network trained for 200 epochs, lr=0.1 ● after 160 epochs lr=0.01 ● at 120 epoch loss prediction module ● does not influence the con weights

  23. Our ideas for Active Learning 1. Remove the loss prediction module and use decoder or VAE. Take to the labelling process samples with the highest reconstruction loss

  24. Our ideas for Active Learning 1. Remove the loss prediction module and use decoder or VAE.

  25. Our ideas for Active Learning 2. We should try to make an adversarial example of the image and choose those which requires the smallest modification to do so. DONE: https://arxiv.org/pdf/1802.09841.pdf

  26. Our ideas for Active Learning 3. We should be like GANs. Train a discriminator to distinguish between labeled and unlabeled dataset. DONE: https://arxiv.org/pdf/1907.06347.pdf

  27. Our ideas for Active Learning 4. The neural network is learning the easy example first. Can be the history of learning a differentiation between labeled and unlabeled datasets?

  28. Our ideas for Active Learning 4. History of learning every 20 epochs History record w a Labeled r Unlabeled Label dataset

  29. Our ideas for Active Learning 4. History of learning Labeled Labeled Unlabeled History record RandomForest dataset Unlabeled 1000 Only unlabeled 100 sampled with highest possibility of being unlabeled

  30. Our ideas for Active Learning 4. History of learning

  31. Our ideas for Active Learning 5. Different moments of History of learning

  32. Our ideas for Active Learning 6. Maybe the NN is after critical point - do not fine-tune

  33. Our ideas for Active Learning 7. Take the history of inner layers

  34. Our ideas for Active Learning 8. Why entropy is so good? Can we be better?

  35. Our ideas for Active Learning 9. Is history even worth something?

  36. Our ideas for Active Learning 9. Is history even worth something?

  37. Our ideas for Active Learning 9. Is history even worth something?

  38. Our ideas for Active Learning 9. Is history even worth something?

  39. Our ideas for Active Learning 9. Is history even worth something?

  40. Our ideas for Active Learning 9. Is history even worth something?

  41. Future plans 1. Investigate ways of finding the dataset outlier. 2. Do more research about history of learning.

  42. Future plans 1. IDEA: use augmentation for checking how the image prediction is sustained through different transformation. Take samples with highest m . y’ y’’ y’’’

  43. Bibliography 1. Yoo, Donggeun, and In So Kweon. "Learning Loss for Active Learning." Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2019. https://arxiv.org/pdf/1905.03677.pdf 2. Ducoffe, Melanie, and Frederic Precioso. "Adversarial active learning for deep networks: a margin based approach." arXiv preprint arXiv:1802.09841 (2018). https://arxiv.org/pdf/1802.09841.pdf 3. Gissin, Daniel, and Shai Shalev-Shwartz. "Discriminative active learning." arXiv preprint arXiv:1907.06347 (2019). https://arxiv.org/pdf/1907.06347.pdf 4. Sener, Ozan, and Silvio Savarese. "Active learning for convolutional neural networks: A core-set approach." arXiv preprint arXiv:1708.00489 (2017). https://arxiv.org/pdf/1708.00489.pdf

Download Presentation
Download Policy: The content available on the website is offered to you 'AS IS' for your personal information and use only. It cannot be commercialized, licensed, or distributed on other websites without prior consent from the author. To download a presentation, simply click this link. If you encounter any difficulties during the download process, it's possible that the publisher has removed the file from their server.

Recommend

Agenda Intro to Active Learning Activity Design Resources for Active Learning Lunch with Active

Agenda Intro to Active Learning Activity Design Resources for Active Learning Lunch with Active

Agenda Intro to Active Learning Activity Design Resources for Active Learning Lunch with Active Learning Veterans Wrap Up Introduction to Active Learning What is active learning? Brainstorm your ideas with your group and generate a

855 views • 42 slides
Multi-Task Active Learning Yi Zhang Outline Active Learning Multi-Task Active Learning

Multi-Task Active Learning Yi Zhang Outline Active Learning Multi-Task Active Learning

Multi-Task Active Learning Yi Zhang Outline Active Learning Multi-Task Active Learning Linguistic Annotations (ACL 08) Image Classification (CVPR 08) Current Work and Discussions Constraint-Driven Active Learning

945 views • 47 slides
Active Learning Passive Learning Active Learning 1. Think 1. Acquisition of knowledge Ability

Active Learning Passive Learning Active Learning 1. Think 1. Acquisition of knowledge Ability

Active Learning Passive Learning Active Learning 1. Think 1. Acquisition of knowledge Ability 2. Decision making 2. Application of knowledge 3. Explain/communicate 1. Deep learning Approach to Uni-directional 2. Interactive Learning

282 views • 5 slides
Learning as Loss Minimization Machine Learning 1 Learning as loss minimization The setup

Learning as Loss Minimization Machine Learning 1 Learning as loss minimization The setup

Learning as Loss Minimization Machine Learning 1 Learning as loss minimization The setup Examples x drawn from a fixed, unknown distribution D Hidden oracle classifier f labels examples We wish to find a hypothesis h that mimics f

294 views • 25 slides
ACTIVE 1. WELCOME and JUMPING IN LEARNING 1. WHAT IS ACTIVE LEARNING? la Carte 1.

ACTIVE 1. WELCOME and JUMPING IN LEARNING 1. WHAT IS ACTIVE LEARNING? la Carte 1.

ACTIVE 1. WELCOME and JUMPING IN LEARNING 1. WHAT IS ACTIVE LEARNING? la Carte 1. CONSTRUCTING our TEACHING CHALLENGES in a low tech 1. ORCHESTRATION environment 1. SHOW-AND-TELL Selma Hamdani Psychology / DALC / Neuroscience

836 views • 50 slides
12.1 Active Learning: A Review When learning, it may be the case that getting the true labels of

12.1 Active Learning: A Review When learning, it may be the case that getting the true labels of

CS/CNS/EE 253: Advanced Topics in Machine Learning Topic: Active Learning Review and Kernel Methods Lecturer: Andreas Krause Scribe: Jonathan Krause Date: Feb. 17, 2010 12.1 Active Learning: A Review When learning, it may be the case that getting

213 views • 5 slides
Predator and Prey: Active Learning Is Social Learning Kenneth Ronkowitz Active Learning

Predator and Prey: Active Learning Is Social Learning Kenneth Ronkowitz Active Learning

Predator and Prey: Active Learning Is Social Learning Kenneth Ronkowitz Active Learning Symposium, Rutgers University, May 15, 2018 Typically part of the curriculum in biology, phrenology, biology, anatomy, craniometry, physiognomy, crime

484 views • 14 slides
10.1 Active Learning Review What questions should be asked in order to collect best data?

10.1 Active Learning Review What questions should be asked in order to collect best data?

CS/CNS/EE 253: Advanced Topics in Machine Learning Topic: Active Learning 2: the myopic version-space shrinking strategy Lecturer: Daniel Golovin Scribe: Pete Trautman Date: 8 February 2010 10.1 Active Learning Review What questions should be

264 views • 4 slides
Active Learning: Rethinking Our Teaching to Promote Deeper Learning Facilitated by Ken Silvestri,

Active Learning: Rethinking Our Teaching to Promote Deeper Learning Facilitated by Ken Silvestri,

Active Learning: Rethinking Our Teaching to Promote Deeper Learning Facilitated by Ken Silvestri, CFE Instructional Designer Adapted for Institute of Technology By Menaka Abraham OVERVIEW Why Active Learning What is Active Learning

542 views • 18 slides
Online Learning with Pairwise Loss Functions Online Learning with Pairwise Loss Functions MLSIG

Online Learning with Pairwise Loss Functions Online Learning with Pairwise Loss Functions MLSIG

Online Learning with Pairwise Loss Functions Online Learning with Pairwise Loss Functions MLSIG Seminar Series, Dept. of CSA, IISc Joint work with B. Sriperumbudur, P. Jain, H. Karnick Purushottam Kar MLO Group, Microsoft Research India Outline

464 views • 28 slides
Using R for Active Learning and Self-assessment in an e-Learning Environment Arnau Mir, Margaret

Using R for Active Learning and Self-assessment in an e-Learning Environment Arnau Mir, Margaret

Using R for Active Learning and Self-assessment in an e-Learning Environment Arnau Mir, Margaret Mir-Juli and Monica J. Ruiz-Mir University of the Balearic Islands Palma de Mallorca, SPAIN Using R for Active Learning and Self-assessment

740 views • 12 slides
Active Learning for Regression: Active Learning for Regression: Algorithms and Applications

Active Learning for Regression: Active Learning for Regression: Algorithms and Applications

LAMDA group, Nanjing University Nov. 5, 2009. Active Learning for Regression: Active Learning for Regression: Algorithms and Applications Algorithms and Applications Masashi Sugiyama Tokyo Institute of Technology sugi@cs.titech.ac.jp

870 views • 64 slides
Course outline Online decision making 1. Statistical active learning 2. Combinatorial

Course outline Online decision making 1. Statistical active learning 2. Combinatorial

Active Learning and Optimized Information Gathering Lecture 3 Reinforcement Learning CS 101.2 Andreas Krause Announcements Homework 1: out tomorrow Due Thu Jan 22 Project Proposal due Tue Jan 27 (start soon!) Office hours Come to

573 views • 23 slides
Using active learning pedagogy to enable the Presentation by Donna Whitehead teaching of

Using active learning pedagogy to enable the Presentation by Donna Whitehead teaching of

Using active learning pedagogy to enable the Presentation by Donna Whitehead teaching of enterprise and entrepreneurial skills Active learning Active learning methodology has become a preferred way to change the traditional teacher

146 views • 11 slides
Active Learning with Active Learning with Model Selection Neil Rubens Sugiyama Lab / Tokyo

Active Learning with Active Learning with Model Selection Neil Rubens Sugiyama Lab / Tokyo

Active Learning with Active Learning with Model Selection Neil Rubens Sugiyama Lab / Tokyo Institute of Technology Active Learning (NLP Motivation) NLP (common scenario) Large amounts of unlabeled data Large amounts of unlabeled

706 views • 36 slides
Approximating Learning Curves for Active-Learning-Driven Annotation Katrin T omanek and Udo Hahn

Approximating Learning Curves for Active-Learning-Driven Annotation Katrin T omanek and Udo Hahn

Approximating Learning Curves for Active-Learning-Driven Annotation Katrin T omanek and Udo Hahn Jena University Language and Information Engineering (JULIE) Lab Agenda Introduction to Active Learning Stopping Conditions

511 views • 19 slides
Machine learning theory Active learning Hamid Beigy Sharif university of technology June 13,

Machine learning theory Active learning Hamid Beigy Sharif university of technology June 13,

Machine learning theory Active learning Hamid Beigy Sharif university of technology June 13, 2020 Table of contents 1. Introduction 2. Active learning 3. Summary 1/18 Introduction Introduction We have studied the passive supervised

406 views • 23 slides
What makes a good lecturer? Active Learning Richard Anderson University of Washington June 30,

What makes a good lecturer? Active Learning Richard Anderson University of Washington June 30,

What makes a good lecturer? Active Learning Richard Anderson University of Washington June 30, 2008 IUCEE: Active Learning 1 June 30, 2008 IUCEE: Active Learning 2 Student Attention Classroom Activities Planned interactions with

359 views • 5 slides
The patients The patients and how that may impact the provider/patient relationship Identify

The patients The patients and how that may impact the provider/patient relationship Identify

Disclosures I have no relevant financial relationships with any companies related to the content of this course. Treating the Millennial A Generationally-Minded Approach to Transition-Aged Patients Taryn Shappell, LCSW 2 9/7/2018 Learning

287 views • 7 slides
FROM EXPERIMENTS WITH SELECTED VERTICAL DOMAINS Prof. Panagiotis Demestichas, University of

FROM EXPERIMENTS WITH SELECTED VERTICAL DOMAINS Prof. Panagiotis Demestichas, University of

EMERGING 5G TECHNOLOGIES AND INSIGHTS FROM EXPERIMENTS WITH SELECTED VERTICAL DOMAINS Prof. Panagiotis Demestichas, University of Piraeus Dr. Andreas Georgakopoulos, WINGS ICT Solutions 5G Summit, Thessaloniki, 11 July 2017 Outline 5G-PPP:

239 views • 12 slides
Active Learning SPiNCOM reading group Sep. 30 th , 2016 Dimitris Berberidis 1 A toy example:

Active Learning SPiNCOM reading group Sep. 30 th , 2016 Dimitris Berberidis 1 A toy example:

Active Learning SPiNCOM reading group Sep. 30 th , 2016 Dimitris Berberidis 1 A toy example: Alien fruits Consider alien fruits of various shapes Train classifier to distinguish safe fruits from dangerous ones Passive learning:

570 views • 25 slides
Learning as Tools for Biosecurity Education Jo L. Husbands Scholar/Senior Project Director

Learning as Tools for Biosecurity Education Jo L. Husbands Scholar/Senior Project Director

Responsible Science and Active Learning as Tools for Biosecurity Education Jo L. Husbands Scholar/Senior Project Director Board on Life Sciences Three Assumptions Education as the foundation for biorisk management/biosecurity Power of

546 views • 8 slides
to promote student conceptual understanding Sahana Murthy IDP in Educational Technology Indian

to promote student conceptual understanding Sahana Murthy IDP in Educational Technology Indian

Peer-Instruction: An active learning strategy to promote student conceptual understanding Sahana Murthy IDP in Educational Technology Indian Institute of Technology Bombay June 17, 2014 ISTE Workshop on Computer Programming June 17, 2014 1

1.03k views • 36 slides
reduced in active learning of agent strategies? Cline Hocquette & Stephen H. Muggleton

reduced in active learning of agent strategies? Cline Hocquette & Stephen H. Muggleton

How much can experimental cost be reduced in active learning of agent strategies? Cline Hocquette & Stephen H. Muggleton Learning agent strategies from observations Experimentation requires energy, time and resources Automated

613 views • 16 slides

More recommend