I have been thinking a lot about this paper about why explainable AI is not ready for patient-level explainability of AI in healthcare: https://t.co/z9FG2jXe7P
— Christoph Molnar (@ChristophMolnar) November 25, 2021
While I would not go as far as calling it a "false hope", there is food for thought in there.
1/n
TransMix: Attend to Mix for Vision Transformers
— AK (@ak92501) November 19, 2021
abs: https://t.co/P1m7LHzOqH pic.twitter.com/PJeURpaWpw
Great paper and thread!
— Andrej Karpathy (@karpathy) November 13, 2021
- 😮that super simple MSE loss works vs. BEiT-style dVAE (multi-modal) cross-entropy
- <3 efficiency of asymmetric encoder/decoder
- 👏detailed training recipes
- +1 v curious about dataset size scaling
- bit of lack of commentary on test-time protocol https://t.co/MQFAvrqBvr
TensorFlow.js found its killer app for Virtual YouTubers https://t.co/lbh6akLMOq
— hardmaru (@hardmaru) November 12, 2021
Dense Unsupervised Learning for Video Segmentation
— AK (@ak92501) November 12, 2021
abs: https://t.co/zKV94Pn7s5
github: https://t.co/8WQyRw3jw9 pic.twitter.com/nST5odpRMU
.@Gradio Demo for AnimeGANv2 Face Portrait v2 now on @huggingface Spaces
— AK (@ak92501) November 6, 2021
demo: https://t.co/ElxylWJGwj
github: https://t.co/w7qvAwmS3m pic.twitter.com/jjnJN0ZlQf
The Deep Learning Devil is in the Details. I love this work from @IrwanBello and collaborators in which they show how training "tricks" improve ~3% absolute accuracy on ImageNet, progress equivalent to years of developments and research!
— Oriol Vinyals (@OriolVinyalsML) November 5, 2021
Paper: https://t.co/oA6MDfq1XB pic.twitter.com/auvM5yrt01
Image-Based CLIP-Guided Essence Transfer
— AK (@ak92501) October 26, 2021
abs: https://t.co/wFGwhJxRCZ
github: https://t.co/38xy9RrC9x
new method creates a blending operator that is optimized to be simultaneously additive in both latent spaces pic.twitter.com/WqKURD8ny8
FlexMatch: Boosting Semi-Supervised Learning with Curriculum Pseudo Labeling
— AK (@ak92501) October 20, 2021
abs: https://t.co/FDZTByplIY
FlexMatch outperforms FixMatch by 14.32% and 24.55% on CIFAR-100 and STL-10 datasets respectively, when there are only 4 labels per class pic.twitter.com/uMYQ171WoL
Boosting Contrastive Self-Supervised Learning with False Negative Cancellation
— AK (@ak92501) October 19, 2021
abs: https://t.co/hrLNic6CRL
github: https://t.co/FwpPAPxAhp pic.twitter.com/BNLRIZImul
Understanding and Improving Robustness of Vision
— AK (@ak92501) October 18, 2021
Transformers through Patch-based Negative
Augmentation
abs: https://t.co/mFJURTVAHN
show that patch-based negative augmentation consistently improves robustness of ViTs across a wide set of ImageNet based robustness benchmarks pic.twitter.com/CUQrtfdzxe
Introducing a minimalist and effective approach for vision language model pre-training that learns a single representation from both visual and language inputs and efficiently leverages scaled datasets to achieve state-of-the-art performance. Learn more ↓ https://t.co/U9DY2CZbqR
— Google AI (@GoogleAI) October 15, 2021
