A day at the beach: Deep learning for a child
https://buff.ly/2H8hYwP
https://buff.ly/2H8hYwP
The Conversation
A day at the beach: Deep learning for a child
Through a play day filled with choices at the beach with supportive adults, unexpected challenges and social experiences all help children to build far more than sand castles.
PHYRE: A New Benchmark for Physical Reasoning
By: Anton Bakhtin, Laurens van der Maaten, Justin Johnson, Laura Gustafson and Ross Girshick. Facebook AI Research : https://research.fb.com/publications/phyre-a-new-benchmark-for-physical-reasoning/
Demo: https://player.phyre.ai
#DeepLearning #Physics #ReinforcementLearning
By: Anton Bakhtin, Laurens van der Maaten, Justin Johnson, Laura Gustafson and Ross Girshick. Facebook AI Research : https://research.fb.com/publications/phyre-a-new-benchmark-for-physical-reasoning/
Demo: https://player.phyre.ai
#DeepLearning #Physics #ReinforcementLearning
Facebook Research
PHYRE: A New Benchmark for Physical Reasoning - Facebook Research
Understanding and reasoning about physics is an important ability of intelligent agents. We develop the PHYRE benchmark for physical reasoning that contains a set of simple classical mechanics puzzles in a 2D physical environment.
Does the brain do backpropagation? CAN Public Lecture - Geoffrey Hinton
One of the best recent talks of Prof. geoffrey hinton
online on computation in the brain. Intriguingly, the proposed relation between the neuron firing rate and the error signal looks quite similar to the Euler-Lagrange equation of motion in Physics.
https://www.youtube.com/watch?v=qIEfJ6OBGj8
@ArtificialIntelligenceArticles
One of the best recent talks of Prof. geoffrey hinton
online on computation in the brain. Intriguingly, the proposed relation between the neuron firing rate and the error signal looks quite similar to the Euler-Lagrange equation of motion in Physics.
https://www.youtube.com/watch?v=qIEfJ6OBGj8
@ArtificialIntelligenceArticles
YouTube
Does the brain do backpropagation? CAN Public Lecture - Geoffrey Hinton - May 21, 2019
Canadian Association for Neuroscience 2019 Public lecture: Geoffrey Hinton
https://can-acn.org/2019-public-lecture-geoffrey-hinton
https://can-acn.org/2019-public-lecture-geoffrey-hinton
Visualizing and Measuring the Geometry of BERT
Coenen et al.: https://arxiv.org/abs/1906.02715
#BERT #NaturalLanguageProcessing #UnsupervisedLearning
Coenen et al.: https://arxiv.org/abs/1906.02715
#BERT #NaturalLanguageProcessing #UnsupervisedLearning
arXiv.org
Visualizing and Measuring the Geometry of BERT
Transformer architectures show significant promise for natural language processing. Given that a single pretrained model can be fine-tuned to perform well on many different tasks, these networks...
An Atari Model Zoo for Analyzing, Visualizing, and Comparing Deep Reinforcement Learning Agents
Such et al.: https://arxiv.org/abs/1812.07069
Code: https://github.com/uber-research/atari-model-zoo
Blog: https://eng.uber.com/atari-zoo-deep-reinforcement-learning/
#ArtificialIntelligence #DeepLearning #ReinforcementLearning
Such et al.: https://arxiv.org/abs/1812.07069
Code: https://github.com/uber-research/atari-model-zoo
Blog: https://eng.uber.com/atari-zoo-deep-reinforcement-learning/
#ArtificialIntelligence #DeepLearning #ReinforcementLearning
Deep Kernel Learning for Clustering
Wu et al.: https://arxiv.org/pdf/1908.03515v1.pdf
#DeepLearning #MachineLearning #NeuralNetworks
Wu et al.: https://arxiv.org/pdf/1908.03515v1.pdf
#DeepLearning #MachineLearning #NeuralNetworks
"One-shot Face Reenactment"
Zhang et al.: https://arxiv.org/abs/1908.03251
Project: https://wywu.github.io/projects/ReenactGAN/OneShotReenact.html
GitHub: https://github.com/bj80heyue/One_Shot_Face_Reenactment
#ArtificialIntelligence #DeepLearning #MachineLearning
Zhang et al.: https://arxiv.org/abs/1908.03251
Project: https://wywu.github.io/projects/ReenactGAN/OneShotReenact.html
GitHub: https://github.com/bj80heyue/One_Shot_Face_Reenactment
#ArtificialIntelligence #DeepLearning #MachineLearning
Creative improvisation is at the root of jazz and science.
https://www.openculture.com/2016/07/the-secret-link-between-jazz-and-physics-how-einstein-coltrane-shared-improvisation-and-intuition-in-common.html
https://www.openculture.com/2016/07/the-secret-link-between-jazz-and-physics-how-einstein-coltrane-shared-improvisation-and-intuition-in-common.html
Open Culture
The Secret Link Between Jazz and Physics: How Einstein & Coltrane Shared Improvisation and Intuition in Common
Scientists need hobbies. The grueling work of navigating complex theory and the politics of academia can get to a person, even one as laid back as Dartmouth professor and astrophysicist Stephon Alexander.
Multi-Agent Manipulation via Locomotion using Hierarchical Sim2Real
Nachum et al.: https://arxiv.org/abs/1908.05224
#Robotics #ArtificialIntelligence #MachineLearning
Nachum et al.: https://arxiv.org/abs/1908.05224
#Robotics #ArtificialIntelligence #MachineLearning
Something really really cool!🙂
#weekend_read
Paper-Title: Multi-Agent Manipulation via Locomotion using Hierarchical Sim2Real #GoogleAI
Link to the paper: https://arxiv.org/pdf/1908.05224.pdf
Link to the videos: https://sites.google.com/view/manipulation-via-locomotion
TL;DR: They have presented successful zero-shot transfer of policies trained in simulation to perform difficult locomotion and manipulation via locomotion tasks. The key to their method is the imposition of hierarchy, which introduces modularity into the domain randomization process and enables the learning of increasingly complex behaviours.
#weekend_read
Paper-Title: Multi-Agent Manipulation via Locomotion using Hierarchical Sim2Real #GoogleAI
Link to the paper: https://arxiv.org/pdf/1908.05224.pdf
Link to the videos: https://sites.google.com/view/manipulation-via-locomotion
TL;DR: They have presented successful zero-shot transfer of policies trained in simulation to perform difficult locomotion and manipulation via locomotion tasks. The key to their method is the imposition of hierarchy, which introduces modularity into the domain randomization process and enables the learning of increasingly complex behaviours.
Google
Multi-Agent Manipulation via Locomotion using Hierarchical Sim2Real
Two D"Kitties co-ordinate to push a heavy block to the target (marked by + signs on the floor)
Best Resources for Getting Started With GANs
https://machinelearningmastery.com/resources-for-getting-started-with-generative-adversarial-networks/
https://machinelearningmastery.com/resources-for-getting-started-with-generative-adversarial-networks/
MachineLearningMastery.com
Best Resources for Getting Started With GANs - MachineLearningMastery.com
Generative Adversarial Networks, or GANs, are a type of deep learning technique for generative modeling. GANs are the techniques behind the startlingly photorealistic generation of human faces, as well as impressive image translation tasks such as photo colorization…
Pytorch Implementation of Autoregressive Language Model
https://github.com/lyeoni/pretraining-for-language-understanding
https://github.com/lyeoni/pretraining-for-language-understanding
GitHub
GitHub - lyeoni/pretraining-for-language-understanding: Pre-training of Language Models for Language Understanding
Pre-training of Language Models for Language Understanding - GitHub - lyeoni/pretraining-for-language-understanding: Pre-training of Language Models for Language Understanding
Unconstrained Monotonic Neural Networks
Antoine Wehenkel and Gilles Louppe : https://arxiv.org/abs/1908.05164
#NeuralNetworks #MachineLearning #NeuralComputing
Antoine Wehenkel and Gilles Louppe : https://arxiv.org/abs/1908.05164
#NeuralNetworks #MachineLearning #NeuralComputing
12 NLP Researchers, Practitioners & Innovators You Should Be Following
Check out this list of NLP researchers, practitioners and innovators you should be following, including academics, practitioners, developers, entrepreneurs, and more.
https://www.kdnuggets.com/2019/08/nlp-researchers-practitioners-innovators-should-follow.html
Check out this list of NLP researchers, practitioners and innovators you should be following, including academics, practitioners, developers, entrepreneurs, and more.
https://www.kdnuggets.com/2019/08/nlp-researchers-practitioners-innovators-should-follow.html
Conditional Neural Processes
Garnelo et al.: https://arxiv.org/abs/1807.01613
1. Repo: https://github.com/deepmind/neural-processes
2. NoteBook: https://github.com/deepmind/neural-processes/blob/master/conditional_neural_process.ipynb
#ArtificialIntelligence #MachineLearning
Garnelo et al.: https://arxiv.org/abs/1807.01613
1. Repo: https://github.com/deepmind/neural-processes
2. NoteBook: https://github.com/deepmind/neural-processes/blob/master/conditional_neural_process.ipynb
#ArtificialIntelligence #MachineLearning
arXiv.org
Conditional Neural Processes
Deep neural networks excel at function approximation, yet they are typically trained from scratch for each new function. On the other hand, Bayesian methods, such as Gaussian Processes (GPs),...
A guide to convolution arithmetic for deep learning
Vincent Dumoulin and Francesco Visin : https://arxiv.org/pdf/1603.07285.pdf
#artificialintelligence #deeplearning #machinelearning
Vincent Dumoulin and Francesco Visin : https://arxiv.org/pdf/1603.07285.pdf
#artificialintelligence #deeplearning #machinelearning
CS231n Convolutional Neural Networks for Visual Recognition
https://cs231n.github.io/neural-networks-3/#loss
https://cs231n.github.io/neural-networks-3/#loss
cs231n.github.io
CS231n Deep Learning for Computer Vision
Course materials and notes for Stanford class CS231n: Deep Learning for Computer Vision.
Sports Matches & Artificial Intelligence
https://www.youtube.com/watch?v=kaslJ-8piSE&feature=youtu.be
https://www.youtube.com/watch?v=kaslJ-8piSE&feature=youtu.be
YouTube
Sports Matches & Artificial Intelligence
#ComputerVision supported by #DeepLearning to help SPORT ANALYTICS Achieving fully automated, without manual operators and wearables, real-time individual pl...
Gitflow - Using git in a right way
https://sharetechlinks.com/post/detail/89/gitflow-using-git-in-a-right-way
https://sharetechlinks.com/post/detail/89/gitflow-using-git-in-a-right-way
Sharetechlinks
Gitflow - Using Git in a right way - Share Tech Links
Git is an open-source distributed version control system that is flexible and easy to use for all kinds of teams, no matter how big or small. To adopt Git in everyday development, a model called Gitflow was introduced by Vincent Driessen to help simplify…
Does Deep Learning Still Need Backpropagation?
https://syncedreview.com/2019/08/14/does-deep-learning-still-need-backpropagation/
https://syncedreview.com/2019/08/14/does-deep-learning-still-need-backpropagation/
Synced
Does Deep Learning Still Need Backpropagation?
Now, researchers from the Victoria University of Wellington School of Engineering and Computer Science have introduced the HSIC (Hilbert-Schemidt independence criterion) bottleneck as an alternativ…
Bayesian Generative Models for Knowledge Transfer in MRI Semantic Segmentation Problems. https://arxiv.org/abs/1908.05480