Excellent Course...Thank You very much...

Actual content starts at 57:13

very nice

Thanks for this amazing lecture visual recognition models. Really appricated!

Thanks so much for making this available!

Thanks for the nice presentation but It could have been better. Maybe you can make a series of short videos to explain these with examples.

Thank you so much ❤❤

Thank you so much ❤❤

Thank you so much for providing this amazing content for free! Your efforts are truly appreciated! ❤❤

Your lecture is amazing, I love how you give so many practical examples!

Many thanks for this useful talk

Hi, thank for this very good tutorial series. May I know if there is an issue in the collab notebook used in this episode? It reports an error on the as_classification_dataset...

Such a great course for continual learning

Thank you so so much for sharing this information for free. 🎉

Thank you so much for this❤❤

17:43 24:01 53:18

0:18 Continual Learning of Object Instances 1:17 Generative Feature Replay for Class-incremental Learning 2:17 A simple Class Decision Balancing for Incremental Learning 3:10 Adaptive Group Sparse Regularization for Continual Learning 4:10 CatNet: Class Incremental 3D ConvNets for Lifelong EgoCentric Gesture Recognition 5:17 Few-shot Image Recognition for UAV Sports Cinematography 6:09 Generalized Class Incremental Learning (my favorite) 7:10 Rehearsal-Free Continual Learning over Small Non-I.I.D Batches 8:10 Continual Reinforcement Learning in 3D Non-stationary Environments 9:10 Reducing catastrophic forgetting with learning on synthetic data 10:13 Continual Learning for Anomaly detection in surveillance videos

This is really a very good well rounded course until this episode. You saved me much time and gave me the needed information about what's going in research of CL instead of spending too much time to discover the field. Thanks a lot.

good

One strategy to rule them all 😂

Can u share this slide? Thank u so much!

Very Great work

🅿🆁🅾🅼🅾🆂🅼

I Love this video

Faultlessness🎇

I like your explanation and examples. But taking too much time in small ideas

Finally the last lecture :D, eager to see more on this course! Thanks!

This is great. I really wish the documentation had more working colab examples/tutorials so we could play around with the whole thing more.

Thank you very much for making this course public!

dope tech!

Hi, great lecture series! I was wondering if reinforcement learning can be considered as continual learning?

awesome work!

What we have learned today: Daenerys Targaryen is the mother of all strategies ;) haha

Great lecture! I like the compass idea too :) Especially to be able to find similar paper later which possibly work on the same challenges and problem types.

Perfect explanation

great！it can help me to know about the direction，which is new ，and puzzle me for a while.

amogus

thanks so much for uploading this course; about to go through it!

Speaker was muted from 0:50 till 4:36 =(

Hi, Thank you very much for such video. Can we add incremental learning (Creme by Keras) in facial emotion recognition?

Thanks for these interesting videos! Just one piece of feedback: the intro is REALLY loud, would be awesome if you can turn it down a bit :)

11:32 Introduction. Vincenzo Lomonaco([email protected]), University of Pisa. 23:23 Continue Learning the challenge. Razvan Pascanu([email protected]), Research Scientist @ DeepMind. 55:18 CONTINUE LEARNING FOR AFFECTIVE ROBOTICS. Hatice Gunes. Reader in Affective Intelligence & Robotics, University of Cambridge, Department of Computer Science & Technology. 1:29:06 Exemplar-Free class-incremental Learning. Joost van de Weijer, Learning and Machine Perception group. Computer Vision Center, Universitat Autonoma de Barcelona. 2:0929 Break. 2:47:02 Continue Learning: Repetition, Reconstruction, and Forgetting. Jim Rehg. 3:22:23 Continual Learning: A story line and a wider scope. Rahaf Aljundi, Toyota Motor Europe. 3:57:06 A Table of Two CILS: The connection between Class Incremental Learning and Class Imbalanced Learning, and Beyond. Chen He, Key Lab of Intelligent Information Processing, Institute of Computing Technology, Chinese Academy of Sciences.

how can i attend these meetup?

started at 13:34

what do you call experience in the library? can you please elaborate.

Great talk Chris and Thanks Continual ai community for opensource-ing this❤️ Chris make very important points.👍

great, thank you :D

Can i get mail id of speaker?

Have you tried running the autoencoder on classes that haven't yet been seen by the model? If most images of "future" classes cannot be well reconstructed, it would definitely rule out the hypothesis that the classes are similarly distributed. At least locally.

you misspelled his name...