Ribana Roscher
Ribana Roscher
  • Видео 23
  • Просмотров 68 641
Explainable machine learning (2022, 5th lecture): Interpret. by backward prop. + concept discovery
Lecture series 'Explainable machine learning' (University of Bonn, winter term 2022)
Topics covered:
- Gradient-based saliency maps
- CAM and Grad-CAM
- Layer-wise relevance propagation
- Testing with concept activation vectors (TCAV)
- Activations space occlusion sensitivity
Slides: uni-bonn.sciebo.de/s/gfz4cdWE57jzRk8
Lecturer: Ribana Roscher
Winter term 2022, University of Bonn
Просмотров: 312

Видео

Explainable machine learning (2022, 4th lecture): Local model-agnostic methods
Просмотров 450Год назад
Lecture series 'Explainable machine learning' (University of Bonn, winter term 2022) Topics covered: - What are local model-agnostic methods? - Shapley values - Occlusion sensitivity maps - Spectral clustering - Randomized input sampling (RISE) - Local interpretable model-agnostic explanations (LIME) Slides: uni-bonn.sciebo.de/s/ieAfb7k9YSlUrTA Lecturer: Ribana Roscher Winter term 2022, Univers...
Explainable machine learning (2022, 3rd lecture): Global model-agnostic methods
Просмотров 606Год назад
Lecture series 'Explainable machine learning' (University of Bonn, winter term 2022) Topics covered: - What are global model-agnostic methods and why are they useful? - Attribution methods and feature importance - Partial dependence plots - Permutation feature importance - Variance feature importance - Prototypes and criticism - Maximum mean discrepancy and witness function Slides: uni-bonn.sci...
Explainable machine learning (2022, 2nd lecture): Looking into a neural network
Просмотров 8902 года назад
Lecture series 'Explainable machine learning' (University of Bonn, winter term 2022) Topics covered: - Visualizing model components (weights, activations) - Deconvnets - Activation maximization Slides: uni-bonn.sciebo.de/s/PdNGOeSb0yM62yI Lecturer: Ribana Roscher Winter term 2022, University of Bonn
Explainable machine learning (2022, 1st lecture): Introduction
Просмотров 1,5 тыс.2 года назад
Lecture series 'Explainable machine learning' (University of Bonn, winter term 2022) Topics covered: - What is explainable machine learning? - Inherently interpretable models: linear regression and decision trees Slides: uni-bonn.sciebo.de/s/dj6OcgHFOJjY97e Lecturer: Ribana Roscher Winter term 2022, University of Bonn
Remote Sensing Image Analysis and Interpretation: Density estimation
Просмотров 1,6 тыс.3 года назад
Fifth lecture in the course 'Remote Sensing Image Analysis and Interpretation' covering the topics of density estimation, expectation maximization algorithm, and discussing the big picture of the whole course (how all topics connect). slides: uni-bonn.sciebo.de/s/048LClIZuAXBMI3 Lecturer: Ribana Roscher Winter term 2020/2021, University of Bonn
Remote Sensing Image Analysis and Interpretation: Classification with Bayes' theorem
Просмотров 1,8 тыс.3 года назад
Fourth lecture in the course 'Remote Sensing Image Analysis and Interpretation' covering the topics of Maximum likelihood estimation, maximum a posteriori estimation and how a classifier is evaluated. slides: uni-bonn.sciebo.de/s/NS4RnVYJ7s68zyF Lecturer: Ribana Roscher Winter term 2020/2021, University of Bonn
Remote Sensing Image Analysis and Interpretation: Feature extraction and image segmentation
Просмотров 4,2 тыс.3 года назад
Third lecture in the course 'Remote Sensing Image Analysis and Interpretation' discussing what kind of features can be extracted from images, why you need to be careful with high dimensions and what image segmentation is. slides: uni-bonn.sciebo.de/s/kNfkCjMo0UObbE0 Lecturer: Ribana Roscher Winter term 2020/2021, University of Bonn
Remote Sensing Image Analysis and Interpretation: Image analysis and interpretation basics
Просмотров 8 тыс.3 года назад
Second lecture in the course 'Remote Sensing Image Analysis and Interpretation' covering the basics of image analysis and interpretation, change detection and classification. slides: uni-bonn.sciebo.de/s/U6K56GFUaYW9FGt Lecturer: Ribana Roscher Winter term 2020/2021, University of Bonn
Remote Sensing Image Analysis and Interpretation: Introduction to Remote Sensing
Просмотров 15 тыс.4 года назад
First lecture in the course 'Remote Sensing Image Analysis and Interpretation' covering the questions 'What is remote sensing' and 'Why is it useful'. slides: uni-bonn.sciebo.de/s/stS0KK4lESzzaOQ Lecturer: Ribana Roscher Winter term 2020/2021, University of Bonn
Automatic Differentiation of Damaged and Unharmed Grapes Using RGB Images and CNNs
Просмотров 3524 года назад
Presentation given by Laura Zabawa at the ECCV 2020 Workshop on Computer Vision Problems in Plant Phenotyping (CVPPP 2020)
SemCity Toulouse: A Benchmark for Building Instance Segmentation in Satellite Images
Просмотров 1,5 тыс.4 года назад
Presentation given by Ribana Roscher at the ISPRS Congress 2020 Benchmark data can be downloaded here: rs.ipb.uni-bonn.de/data/ Roscher, R., Volpi, M., Mallet, C., Drees, L., & Wegner, J. D. (2020). SemCity Toulouse: A Benchmark for Building Instance Segmentation in Satellite Images. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 5, 109-116.
Explain it to me - Facing Remote Sensing Challenges in the Bio- and Geosciences with Explainable ML
Просмотров 2254 года назад
Presentation given by Ribana Roscher at the ISPRS Congress 2020 Roscher, R., Bohn, B., Duarte, M. F., & Garcke, J. (2020). Explain it to me - Facing Remote Sensing Challenges in the Bio- and Geosciences with Explainable Machine Learning. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 3, 817-824.
Learning with Real-World and Artificial Data for Improved Vehicle Detection in Aerial Imagery
Просмотров 2334 года назад
Presentation given by Immanuel Weber at the ISPRS Congress 2020 Weber, I., Bongartz, J., & Roscher, R. (2020). Learning with Real-World and Artificial Data for Improved Vehicle Detection in Aerial Imagery. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2, 917-924.
Multi-Modal Deep Learning with Sentinel-3 Observations for the Detection of Oceanic Internal Waves
Просмотров 2774 года назад
Presentation given by Lukas Drees at the ISPRS Congress 2020 Drees, L., Kusche, J., & Roscher, R. (2020). Multi-Modal Deep Learning with Sentinel-3 Observations for the Detection of Oceanic Internal Waves. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2, 813-820.
What Identifies a Whale by its Fluke? On the Benefit of Interpretable ML for Whale Identification
Просмотров 2894 года назад
What Identifies a Whale by its Fluke? On the Benefit of Interpretable ML for Whale Identification
Was macht Ihr eigentlich in der Fernerkundung in Bonn?
Просмотров 6254 года назад
Was macht Ihr eigentlich in der Fernerkundung in Bonn?
Advanced Machine Learning for Remote Sensing: Convolutional and Recurrent Neural Networks
Просмотров 3,4 тыс.4 года назад
Advanced Machine Learning for Remote Sensing: Convolutional and Recurrent Neural Networks
Advanced Machine Learning for Remote Sensing: Train neural networks
Просмотров 2 тыс.4 года назад
Advanced Machine Learning for Remote Sensing: Train neural networks
Advanced Machine Learning for Remote Sensing: Representation learning
Просмотров 3,6 тыс.4 года назад
Advanced Machine Learning for Remote Sensing: Representation learning
Advanced Machine Learning for Remote Sensing: Neural Networks
Просмотров 4 тыс.4 года назад
Advanced Machine Learning for Remote Sensing: Neural Networks
Advanced Machine Learning for Remote Sensing: Basics
Просмотров 8 тыс.4 года назад
Advanced Machine Learning for Remote Sensing: Basics
Advanced Machine Learning for Remote Sensing: Welcome
Просмотров 9 тыс.4 года назад
Advanced Machine Learning for Remote Sensing: Welcome

Комментарии

  • @yogeshkumarregar3040
    @yogeshkumarregar3040 3 месяца назад

    Feature Extraction from Remote Sensing High Resolution Data using AI/ML (Ex-High Tension tower, windmill, electric substation, Brick Kiln, farmbunds)

  • @avinashkafle8542
    @avinashkafle8542 4 месяца назад

    thanks much

  • @23win
    @23win 7 месяцев назад

    Hello, I followed your video on remote sensing basics and image classification in the context of GIS with great interest. I find your explanations particularly clear and instructive. While watching this video, a question came to mind: do you think it would be possible to find a thesis topic with a physics emphasis that could be applied in the field of GIS, especially in the context of processing images obtained through remote sensing, or even in remote sensing itself? Although I understand that currently mathematics play a crucial role in many GIS researches, I am convinced that physical principles could also offer interesting perspectives. For example, applying the laws of physics to model light-matter interactions in satellite images, or using advanced physical concepts to improve image classification algorithms. The integration of physical models into geographic information systems for a better understanding of environmental phenomena, or even the use of concepts from quantum physics (qbits - cub sats .....)/statistical physics for innovative applications in the field of geomatics? I would be very interested in your thoughts on this topic. Do you think there are opportunities to further explore this intersection between physics and GIS in the context of academic research, with possible opportunities to conduct experiments in a laboratory, for example, to improve acquisition methods (such as LIDAR)? Thank you again for the time you devote to sharing your knowledge and experience on your RUclips channel. Best regards, AOUINATOU OUSSAMA

  • @shuangwang9886
    @shuangwang9886 8 месяцев назад

    Your explanation is very helpful and excellent. Thank you so much!

  • @praneeljoshi8803
    @praneeljoshi8803 Год назад

    The idea that a NN is not a regressor or classified but a representation learner was profound to me! As a math major, this is the first time ML and NN has been explained in a way that is satisfying and intuitive

  • @Fan-vk9gx
    @Fan-vk9gx Год назад

    This is a really really awesome tutorial! I appreciate your efforts in making such amazing lessons. These lessons provide not only knowledge but also a bunch of organized theories of remote sensing. Thank you very much!

  • @danielsalim5643
    @danielsalim5643 Год назад

    Thanks

  • @Faycal59
    @Faycal59 Год назад

    Thank you for puting the lecture here !!!

  • @Faycal59
    @Faycal59 Год назад

    Thank you

  • @martinder3199
    @martinder3199 Год назад

    Is there a way to get the course material/data to complete it on our own?

    • @ribanaroscher_datascience
      @ribanaroscher_datascience Год назад

      The links to the slides are in the caption of the videos and the exercises with solutions you can find here: uni-bonn.sciebo.de/s/a85mundzq8zC1Xb .

  • @mahermohaisen6302
    @mahermohaisen6302 Год назад

    Thank you for sharing this course, But we need the practical part for more understanding

  • @MrMuhammad2010
    @MrMuhammad2010 Год назад

    Another amazing course. Thanks for sharing your courses for free on RUclips.

  • @MrMuhammad2010
    @MrMuhammad2010 Год назад

    Thanks for sharing. Quite useful.

  • @MrMuhammad2010
    @MrMuhammad2010 Год назад

    Thanks a lot. Very helpful and easy to follow for beginners.

  • @tahjudilwitra9798
    @tahjudilwitra9798 Год назад

    Thanks so much for uploading this lecture. Coincide with I am learning at the moment about XAI for Earth Observation. Hopefully can get a PhD opportunity in this field.

  • @gegovski5360
    @gegovski5360 2 года назад

    Great presentation with many useful information!

  • @eranfeit
    @eranfeit 2 года назад

    Nice. You are welcome to watch my playlist : ruclips.net/p/PLdkryDe59y4Ze9_12JhWu3cs-lOGYwYeD Eran

  • @korbi2275
    @korbi2275 2 года назад

    Thanks a lot for uploading your lecture !! Not currently a student but considering to do a masters in GIS :)

  • @rudeprover
    @rudeprover 2 года назад

    Thank you sooooooooooo much for providing this valuable resource!! It means alot; i will make sure to share it with my friends.

  • @Math2153
    @Math2153 2 года назад

    Thanks a lot for simplified tutorial. Hope for the practical tutorial in future.

  • @balwezrostamhamad3941
    @balwezrostamhamad3941 2 года назад

    it great and useful webinar ,you are going to continue🙂

  • @johnowusukonduah2305
    @johnowusukonduah2305 2 года назад

    Please can you by special request do a practical video on how to implement these in the python language

  • @murtalabadamasi2236
    @murtalabadamasi2236 2 года назад

    This is quite interesting and informative. Thanks for providing this stuff

  • @teutazeka
    @teutazeka 2 года назад

    EXCELLENT LECTURE!!

  • @chunnongli6062
    @chunnongli6062 2 года назад

    you are so great!!! your video is very inspiring!

  • @aminathinanabdulmuhsin1097
    @aminathinanabdulmuhsin1097 2 года назад

    are these basics applicable for both GIS and QGIS?

  • @livemapping3769
    @livemapping3769 2 года назад

    Thank you for sharing this kind of video its more informative and covers the basics of AI and DL

  • @yeozero1101
    @yeozero1101 2 года назад

    Excellent lecture! Thanks for sharing .... :)

  • @zebertuberelaxingcoolandna6758
    @zebertuberelaxingcoolandna6758 2 года назад

    It excellent lecture, I thank you

  • @benchalgomourad3781
    @benchalgomourad3781 2 года назад

    Pefectly explained. congratulations Ribana

  • @elinjoroge4266
    @elinjoroge4266 2 года назад

    can i get the presentation slides or a pdf file of the whole tutorial? nice tutorial i like it

  • @rajitgupta9181
    @rajitgupta9181 3 года назад

    very informative. Thank you alot

  • @gezahagnnegash9740
    @gezahagnnegash9740 3 года назад

    Thanks dear Ribana, it's helpful!

  • @tkasetka
    @tkasetka 3 года назад

    Very nice lecture. Can I share your show your RUclipss to my students?

  • @diegoernestofarfan8120
    @diegoernestofarfan8120 3 года назад

    Muchísimas gracias!

  • @ryano4834
    @ryano4834 3 года назад

    Hi. The tanh activation formula seem to have a problem

    • @ribanaroscher_datascience
      @ribanaroscher_datascience 3 года назад

      Thank you for pointing this out! Yes, the exp in the denominator must be summed, not subtracted.

    • @ryano4834
      @ryano4834 3 года назад

      You're welcome ☺️. I really enjoy your videos. Thanks for the resource

  • @ansamalzubaidy399
    @ansamalzubaidy399 3 года назад

    thank you please can explain about extraction feature of offline signature

  • @samjut3718
    @samjut3718 3 года назад

    Please, how do create remotely sensed image patches for training. Thanks

  • @samjut3718
    @samjut3718 3 года назад

    Very grateful

  • @JOSEREYQJOYNO
    @JOSEREYQJOYNO 3 года назад

    Thank you for this.

  • @JOSEREYQJOYNO
    @JOSEREYQJOYNO 3 года назад

    Thank you for uploading your file. Very useful in my class in remote sensing.

  • @JOSEREYQJOYNO
    @JOSEREYQJOYNO 3 года назад

    Thank you for uploading your file about the discussion of remote sensing image analysis and interpretation. Very informative!

  • @XinhLe
    @XinhLe 3 года назад

    Looking for practical sections. Love the series!

  • @XinhLe
    @XinhLe 3 года назад

    very very useful

  • @ashenafiyazew1666
    @ashenafiyazew1666 3 года назад

    Thankyou for your valuable lecture and I would like to ask you help about how to map a mineral at a specific area through satellite image interpretation before travelling to site as i am exploration geologist

  • @stanitebros
    @stanitebros 3 года назад

    Very Simple Explanation. Thank You

  • @w1232800
    @w1232800 3 года назад

    Very useful lecture for the beginners! Nicely explained in an easy language. Thank you so much for sharing!

  • @utkarshbhardwaj1191
    @utkarshbhardwaj1191 3 года назад

    Hey is your code open source?

  • @mengxiwang4397
    @mengxiwang4397 3 года назад

    Machine learning in forest remote sensing, +1, please

  • @TamNguyen-oz2io
    @TamNguyen-oz2io 3 года назад

    Thank you so much