Playing Atari with deep reinforcement learning –’s approach

From countering an invasion of aliens to demolishing a wall with a ball – AI outperforms humans after just 20 minutes of training.

Spot the flaw – visual quality control in manufacturing

Harnessing the power of AI’s image recognition and deep learning may significantly reduce the cost of visual quality control.

Artificial intelligence imagining and reasoning about the future ML team has been working with Google Brain on helping AI imagine and reason about the future. They started from optimizing TensorFlow’s infrastructure for reinforcement learning and moved to end-to-end training of AI entirely on Google’s newest Cloud TPUs.

Starting deep learning hands-on: image classification on CIFAR-10

Tired of overly theoretical introductions to deep learning? Experiment hands-on with CIFAR-10 image classification with Keras by running code in Neptune.

Image classification sample solution for Kaggle competition

We’ve open sourced image classification sample solution that lets data scientists start competing in the currently running Kaggle Cdiscount competition.

Logo detection and brand visibility analytics

Monitoring brand visibility is an important business problem. We describe our solution for logo detection and visibility analytics with deep learning.

Using deep learning for Single Image Super Resolution

We apply three different deep learning models to reproduce state-of-the-art results in single image super resolution.

Fall 2017 release – launching Neptune 2.1 today!

We’re thrilled today to announce the latest version of Neptune: Machine Learning Lab. This release will allow data scientists using Neptune to take some giant steps forward. Here we take a quick look at each of them.

Solving Atari games with Distributed Reinforcement Learning

We present the result of research conducted at, that focuses on distributing a reinforcement learning algorithm to train on a large CPU cluster

Human log loss for image classification

Deep learning vs human perception: creating a log loss benchmark for industrial & medical image classification problems (e.g. cancer screening).