Machine Learning From Scratch

GitHub : https://github.com/eriklindernoren/ML-From-Scratch

#artificialintelligence #deeplearning #machinelearning

Startup Built A Treasure Trove Of Crop Data By Putting A.I. In The Hands Of Farmers
https://bit.ly/2ZVat3x
plantix.net App for farmers, agricultural workers & plant lovers to increase farming productivity

Today Berlin-based Peat has 25 employees and has raised close to $5 million in venture-capital funding, with their latest Series A round led by London's Index Ventures in December 2017.

Strey’s app, if it can continue to go viral across the world of smallholding farmers in the developing world, shows how the increasing ubiquity of AI, or machine-learning tools, can turn reams of data into something valuable.

Image recognition as a form of AI once skirted the realm of sci-fi, but today thanks to off-the-shelf machine-learning libraries that are freely available from Google, Microsoft and even Amazon and Facebook, the act of “using AI” is becoming commoditized in favor of unique, hard-to-reach data like Peat’s.

Peat uses Google’s TensorFlow software library for its image-recognition tool, notes Strey. “Everyone can do this,” she says of her tool. “What we think is the most powerful in the end, is creating insights out of the data we get from our users.”

Detection of Accounting Anomalies in the Latent Space using Adversarial Autoencoder Neural Networks

Schreyer et al.: https://arxiv.org/abs/1908.00734

#MachineLearning #StatisticalFinance #NeuralNetworks

"Gauge Equivariant Convolutional Networks and the Icosahedral CNN"
Taco S. Cohen, Maurice Weiler, Berkay Kicanaoglu, Max Welling : https://arxiv.org/abs/1902.04615
#ArtificialIntelligence #DeepLearning #MachineLearning

I have created a Web Application for detecting plant diseases.

Using Fast.ai which sits on top of Pytorch with Resnet34 pre-trained model!

Used Plant Village Dataset, Trained it on Google Cloud Platform and deployed it on AWS Elastic Beans!

Accuracy 99.654%

Fork on GitHub: https://github.com/imskr/Plant_Disease_Detection

Editing Text in the Wild. arxiv.org/abs/1908.03047

Semantic Estimation of 3D Body Shape and Pose using Minimal Cameras. arxiv.org/abs/1908.03030

One Model To Rule Them All. arxiv.org/abs/1908.03015

Feature selection of neural networks is skewed towards the less abstract cue. arxiv.org/abs/1908.03000

Ten Trending Academic Papers on the Future of Computer Vision
https://hackernoon.com/top-10-papers-you-shouldnt-miss-from-cvpr-2019-deepfake-facial-recognition-reconstruction-and-more-d5ly3q1w

Music Transformer ( Huang et al, Google Brain, ICLR2019 ) re-implementation repository
github : https://github.com/jason9693/MusicTransformer-tensorflow2.0
library : Tensorflow2.0 (beta)
training env : v100 x 1GPU
paper : https://arxiv.org/abs/1809.04281
Google Magenta blog : https://magenta.tensorflow.org/music-transformer
generated sample demo from this repo : https://www.youtube.com/playlist?list=PLVopZAnUrGWrbIkLGB3bz5nitWThIueS2

For more details, you can see in README.md

Thank you.

ExtremeC3Net: Extreme Lightweight Portrait Segmentation Networks using Advanced C3-modules. arxiv.org/abs/1908.03093

Sim-to-Real Learning for Casualty Detection from Ground Projected Point Cloud Data. arxiv.org/abs/1908.03057

What goes around comes around: Cycle-Consistency-based Short-Term Motion Prediction for A... arxiv.org/abs/1908.03055

Editing Text in the Wild. arxiv.org/abs/1908.03047

CGI faces will soon be indistinguishable from real ones. Here’s how
https://www.digitaltrends.com/cool-tech/cubic-motion-scanning-technology/