We introduce a Classical Japanese dataset called Kuzushiji-MNIST, a drop-in replacement for MNIST, plus 2 other datasets. In this work, we also try more interesting tasks like domain transfer from old Kanji to new Kanji.

Paper: https://lnkd.in/gw-b7Ag
Dataset: https://lnkd.in/gJ9RCHv

✴️ @AI_Python_EN
🗣 @AI_Python_Arxiv
❇️ @AI_Python

Google launches new search engine to help scientists find the datasets they need.

#DataSet

🌎 Link Review


✴️ @AI_Python_EN
🗣 @AI_Python_Arxiv
❇️ @AI_Python

Here is a Deep Graph Library, a Python Package For Graph Neural Networks

The MXNet team and the Amazon Web Services AI lab recently teamed up with New York University / NYU Shanghai to announce Deep Graph Library (DGL), a Python package that provides easy implementations of GNNs research

Click the link to install DGL. :: https://lnkd.in/dmXG8XZ

Web page: https://lnkd.in/dmXG8XZ
Github page: https://lnkd.in/dzG5hrT

#deeplearning #artificialintelligence #machinelearning #gnn

✴️ @AI_Python_EN
🗣 @AI_Python_Arxiv
❇️ @AI_Python

Building a text classification model with TensorFlow Hub and Estimators

🌎 http://bit.ly/2PjbMU7

#AI #MachineLearning #DeepLearning #DataScience


✴️ @AI_Python_EN
🗣 @AI_Python_Arxiv
❇️ @AI_Python

#deeplearning
An educational resource to help anyone learn deep reinforcement learning by OPENAI


https://github.com/openai/spinningup


✴️ @AI_Python_EN
🗣 @AI_Python_Arxiv
❇️ @AI_Python

The 20 core Data Science projects every business should implement.

Link >> https://buff.ly/2i1jIO0

#DataScience #AI #DigitalTransformation

✴️ @AI_Python_EN
🗣 @AI_Python_Arxiv
❇️ @AI_Python

The Future of Artificial Intelligence - The Future of Things Read more here:

https://ift.tt/2PpCVo7

#ArtificialIntelligence #AI #DataScience #MachineLearning #BigData #DeepLearning #NLP #Robots #IoT

✴️ @AI_Python_EN
🗣 @AI_Python_Arxiv
❇️ @AI_Python

How To Run SQL queries in phpMyAdmin

https://goo.gl/QcD614


#Database #Java #javascript #programming #coding #sqlhelp #backend #mysql


✴️ @AI_Python_EN
🗣 @AI_Python_Arxiv
❇️ @AI_Python

talk at Interpretability and Robustness in Audio, Speech, and Language (IRASL) Workshop at NeurIPS2018 are now available online: "Deep Within-Class Covariance Analysis for Robust Deep Audio Representation Learning"

#neurips2018 #neurips #irasl

🌎 Link



✴️ @AI_Python_EN
🗣 @AI_Python_Arxiv
❇️ @AI_Python

Human Centered AI Initiative: a personal vision of how #neuroscience #psychology #ai #physics #mathematics and other fields can work together to both understand biological intelligence and create artificial intelligence! https://hai.stanford.edu/news/the_intertwined_quest_for_understanding_biological_intelligence_and_creating_artificial_intelligence/

✴️ @AI_Python_EN
🗣 @AI_Python_Arxiv
❇️ @AI_Python

poster on "Leveraging machine intelligence on diagnosing UTI" at workshop in #NeurIPS2018.

✴️ @AI_Python_EN
🗣 @AI_Python_Arxiv
❇️ @AI_Python

Thoughtful and succinct explanation about why NIPS was a problematic name for a machine learning conference -- from Jeff Dean ect. at the #neurips2018 workshop for critiquing and correcting trends in ML.

https://www.dropbox.com/s/sv9qcfnv42zbmib/CRACT_2018_paper_35.pdf?dl=0

✴️ @AI_Python_EN
🗣 @AI_Python_Arxiv
❇️ @AI_Python

ICYMI — our poster on Latent Embedding Optimization (LEO) at the #NeurIPS2018 meta-Learning workshop earlier today. We combine a latent generative model of parameters with MAML in the latent space. Our paper demonstrates SOTA results on few-shot classification benchmarks.

✴️ @AI_Python_EN
🗣 @AI_Python_Arxiv
❇️ @AI_Python

Have you heard of #Tube #CNN ?

Object or human detection in video is crucial for many applications.

It can also have useful applications such as in repetitive #manufacturing tasks to monitor and proactively prevent catastrophes.

Compared to images, video provides additional cues which can help to disambiguate the detection problem.

Here the authors attempt to learn discriminative models for the temporal evolution of object appearance and to use such models for object detection.

They introduced space-time tubes corresponding to temporal sequences of bounding boxes. They propose a TPN network where two CNN architectures for generating and classifying tubes, respectively, this helps maximize object recall.

The Tube-CNN then implements a tube-level object detector in the video. Our method improves state of the art on two large-scale datasets for object detection in video: HollywoodHeads and ImageNet VID. Tube models show particular advantages in difficult dynamic scenes.

Link to paper: https://lnkd.in/d3DW5Qe
Pytorch implementation of Tube-CNN : https://lnkd.in/dzxGdtt

#deeplearning #CNN #machinelearning #videoanalytics

✴️ @AI_Python_EN
🗣 @AI_Python_Arxiv
❇️ @AI_Python

Slides from Pieter Abbeel's talk at #NeurIPS2018 workshop on RL under Partial Observability:

https://lnkd.in/eFHdb9d

#NeurIPS #ReinforcementLearning

✴️ @AI_Python_EN
🗣 @AI_Python_Arxiv
❇️ @AI_Python

Basic protocols in quantum reinforcement learning with superconducting circuits

https://www.nature.com/articles/s41598-017-01711-

✴️ @AI_Python_EN
🗣 @AI_Python_Arxiv
❇️ @AI_Python

Progress-in-Quantum-Reinforcement-Learning

http://qtml2017.di.univr.it/resources/Slides/Progress-in-Quantum-Reinforcement-Learning.pdf


✴️ @AI_Python_EN
🗣 @AI_Python_Arxiv
❇️ @AI_Python

R in Pharmacy? Honestly, I started losing any hope to see the pharmacy turning (even slowly!) towards R. A little has been happening since 1976, when the S (father of R), was born. And even when S, then R, gained the status of an industry standard in widely understood bio-sciences, it has never happened in pharmacy, namely in clinical research. This was -and still is- the kingdom exclusively reigned by SAS The King (with a low % of "supporters", including R).

It was a big shame, but, what could have been done against long years of spread myths, doubt, uncertainty and negative attitude?

Well, this is not that everything was right about R! Serious topics still have to be addressed, including:
1) numerical validation (ideally free, coordinated by, say, R Consortium),
2) support for CDISC-related processes,
3) metadata layer (SAS format/informat),
There are more topics, yet there's no place for details.

And then, about 5 years ago, something started changing. Slowly. More and more top-pharma companies (even FDA!) started talking about their use of R publicly, some even contributed (e.g. Merck's gsDesign tool).

Today I'd like to share with you the news: a new initiative by R Consortium - the "R in Pharma" project. http://rinpharma.com/

#R #statistics

✴️ @AI_Python_EN
🗣 @AI_Python_Arxiv
❇️ @AI_Python

"The Matrix Calculus You Need For Deep Learning"

By Terence Parr and Jeremy Howard : https://lnkd.in/dC5MqZM

#100DaysOfMLCode #ArtificialIntelligence #BigData #DeepLearning #MachineLearning

✴️ @AI_Python_EN
🗣 @AI_Python_Arxiv
❇️ @AI_Python

Deeper peer-reviewed dive into how AlphaZero learns to play Chess, Shoji, & Go, beating previous world-class AIs in 4, 2, & 30 hours respectively. Generalizes to potentially any perfect information game. Science

Paper: http://science.sciencemag.org/content/362/6419/1140 …
Blog:

✴️ @AI_Python_EN
🗣 @AI_Python_Arxiv
❇️ @AI_Python