Copy
Twitter
Machine Learning Weekly Digest.
Welcome to this week of the Best of Machine Learning Digest. In this weekly newsletter, we resurface some of the best resources in Machine Learning posted in the past week. This time, we've gotten 52 submissions, including 2 papers.
This newsletter is sponsored by no one ;). Let's change that.

Papers

This week, 2 Papers were posted on Best of ML. In the following, we're showing you the Top 2 posts of this week.
Suphx: Mastering Mahjong with Deep Reinforcement Learning
 
Artificial Intelligence (AI) has achieved great success in many domains, and game AI is widely regarded as its beachhead since the dawn of AI. In recent years, studies on game AI have gradually evolved from relatively simple environments (e.g., perfect-information games such as Go, chess, shogi or two-player imperfect-information games such as heads-up Texas hold'em) to more complex ones (e.g., multi-player imperfect-information games such as multi-player Texas hold'em and StartCraft II). Mahjong is a popular multi-player imperfect-information game worldwide but very challenging for AI research due to its complex playing/scoring rules and rich hidden information. We design an AI for Mahjong, named Suphx, based on deep reinforcement learning with some newly introduced techniques including global reward prediction, oracle guiding, and run-time policy adaptation. Suphx has demonstrated stronger performance than most top human players in terms of stable rank and is rated above 99.99% of all the officially ranked human players in the Tenhou platform. This is the first time that a computer program outperforms most top human players in Mahjong.
 
GANSpace: Discovering Interpretable GAN Controls
 
This paper describes a simple technique to analyze Generative Adversarial Networks (GANs) and create interpretable controls for image synthesis, such as change of viewpoint, aging, lighting, and time of day. We identify important latent directions based on Principal Components Analysis (PCA) applied in activation space. Then, we show that interpretable edits can be defined based on layer-wise application of these edit directions. Moreover, we show that BigGAN can be controlled with layer-wise inputs in a StyleGAN-like manner. A user may identify a large number of interpretable controls with these mechanisms. We demonstrate results on GANs from various datasets.
 

Projects

This week, 18 Projects were posted on Best of ML. In the following, we're showing you the Top 3 posts of this week.
Looking for Inception V3 parallel evaluation in TensorFlow Slim
 
I am using the TensorFlow slim api and the inception v3 model and I have a computer with 8v100 gpus to train the model on. However, Google only provides the code for evaluating the model on a single gpu, which means I will be wasting a ton of resources if I try evaluating the model on my current computer.
 
Mimicry: PyTorch library for reproducibility in GAN research.
 
I've recently built Mimicry, a PyTorch library for GANs which I hope can make GAN research findings more reproducible.
 
YouTube channel that introduces ML through anime
 
I just recently started doing a YouTube project myself where I use anime related stuff hoping to introduce ML and AI aiming to a wider audience and to spread the improvement and potential of ML in the medium.
 






This email was sent to <<Email Address>>
why did I get this?    unsubscribe from this list    update subscription preferences
Monn Ventures · Winterthurerstrasse 649 · Zürich 8051 · Switzerland

Email Marketing Powered by Mailchimp