"Likelihood Ratios for Out-of-Distribution Detection"
Ren et al.: https://arxiv.org/abs/1906.02845
#artificialintelligence #deeplearning #machinelearning

TensorFlow Model Optimization Toolkit — Post-Training Integer Quantization
Blog by TensorFlow: https://medium.com/tensorflow/tensorflow-model-optimization-toolkit-post-training-integer-quantization-b4964a1ea9ba
#tensorflow #artificialintelligence#naturallanguageprocessing

Invertible Residual Networks
Official Pytorch implementation of i-ResNets.
By Jorn Jacobsen: https://github.com/jhjacobsen/invertible-resnet
#pytorch #technology #machinelearning

"Deep Learning for Cognitive Neuroscience" https://arxiv.org/abs/1903.01458
@ArtificialIntelligenceArticles

SLIDES

Evaluating Deep Generative Models on Out-of-Distribution Inputs

Eric NalisnickOxCSML Seminar 31.5.19

https://people.ds.cam.ac.uk/etn22/nalisnick_OxCSML_talk.pdf

SLIDES
ICML Tutorial on Population-Based Methods for Training Deep Neural Networks: Novelty Search, Quality Diversity, Open-Ended Search Algorithms, & Indirect Encoding

https://www.cs.uwyo.edu/~jeffclune/share/2019_06_10_ICML_Tutorial.pdf

"Machine learning approach for low-dose CT imaging yields superior results" via Nature Machine Intelligence


https://m.phys.org/news/2019-06-machine-approach-low-dose-ct-imaging.html

Facebook's machine learning system MelNet generated this AI voice clone of Bill Gates. Others voice clones can be heard here: https://audio-samples.github.io under the heading “Selected Speakers.”

ICML 2019 | Google, ETH Zurich, MPI-IS, Cambridge & PROWLER.io Share Best Paper Honours

https://medium.com/syncedreview/icml-2019-google-eth-zurich-mpi-is-cambridge-prowler-io-share-best-paper-honours-4aeabd5c9fc8

Energy-Based Models applied to the detection of machine-generated text, as opposed to human-produced text.
Brought to you from FAIR.
https://arxiv.org/abs/1906.03351

Are you really into object recognition, but you are sick of looking at 2D boxes and 2D masks? Let's play with 3D shapes!
We build on Mask R-CNN and extend it to infer 3D meshes. Given an input image, we detect all objects, infer their 2D instance boxes and masks as well as their 3D object shapes, all end-to-end! Naturally, we call our approach Mesh R-CNN :D

Paper: https://arxiv.org/abs/1906.02739
Joint work with Jitendra Malik and Justin Johnson

P.S. This task is really hard!

Here are the COMPLETE Lecture notes on Professor Andrew Ng's
Stanford Machine Learning Lecture: https://www.holehouse.org/mlclass/

Tons of presentations from Embedded Vision Summit 2019
https://www.embedded-vision.com/platinum-members/embedded-vision-alliance/embedded-vision-training/downloads/pages/may-2019-summit-slides

Recent paper from CVPR'19 paper about neural free-viewpoint rendering of human avatars without reconstructing geometry

https://www.profillic.com/paper/arxiv:1905.08776

Apparently, GANs are used to create fake profile pictures on LinkedIn for international industrial espionage.

https://apnews.com/bc2f19097a4c4fffaa00de6770b8a60d

What is the fuss about TensorFuzz?
It is the fun automated software “testing” for neural networks,
adapting traditional coverage guided fuzzing techniques.
Run #TensorFuzz to take your test coverage to levels other methods cannot reach (e.g. activation coverage, not just class coverage)

Great work by Augustus Odena and @Ian Goodfellow
Join the #ICML2019 talk at 9:40am today. Grand Ballroom

Read at https://arxiv.org/pdf/1807.10875.pdf
Code: https://github.com/brain-research/tensorfuzz

Shapes and Context:
In-the-wild Image Synthesis & Manipulation

https://www.cs.cmu.edu/~aayushb/OpenShapes/

Few-Shot Adversarial Learning of Realistic Neural Talking Head Models

paper — arxiv
https://arxiv.org/pdf/1905.08233.pdf

video — youtube
https://www.youtube.com/watch?v=p1b5aiTrGzY

Best Deep Learning Research of 2019 So Far

https://medium.com/@ODSC/best-deep-learning-research-of-2019-so-far-7bea0ed22e38