If you enjoy the newsletter, please consider sharing it on Twitter, Facebook, etc! Really appreciate
|
|
January 29 · Issue #76 · View online
The Wild Week in AI is a weekly AI & Deep Learning newsletter curated by @.
|
|
If you enjoy the newsletter, please consider sharing it on Twitter, Facebook, etc! Really appreciate the support :)
|
|
|
Google and Facebook investing in French AI
Google decided to set up a new AI research lab in France, and Facebook announced a €10M investment in FAIR-Paris, Facebook’s European AI research lab. Is France catching up to Canada?
|
Facebook AI chief Yann LeCun moves into pure research role
Yann LeCun, the head of Facebook’s internal artificial intelligence research division, is stepping down from his role to take on a more dedicated research position as chief AI scientist. He will be handing his position over to Jérôme Pesenti, who ran British AI startup Benevolent and prior to that was the chief technology officer of IBM’s Big Data group.
|
MIT researchers make progress in "neuromorphic computing"
Researchers in the emerging field of “neuromorphic computing” have attempted to design computer chips that work like the human brain. Instead of carrying out computations based on binary, on/off signaling, like digital chips do today.
|
TensorFlow 1.5 Release (with eager execution)
Most notably, “Eager Execution” for TensorFlow is now available as a preview. This allows you can execute TensorFlow operations immediately as they are called from Python.
|
|
Faster R-CNN: Down the rabbit hole of modern object detection
An in-depth walkthrough of Faster R-CNN, starting with a higher level overview, and then going over the details for each of the components.
|
How to build your own AlphaZero AI using Python and Keras
The codebase contains a replica of the AlphaZero methodology, built in Python and Keras. Gain a deeper understanding of how AlphaZero works and adapt the code to plug in new games.
|
The philosophical argument for using ROC curves
Focused on, but not limited to, medical AI, this post answers the question of what performance testing is for, and what the features of a good performance metric are.
|
|
FAIR's research platform for object detection research
Detectron is Facebook AI Research’s software system that implements state-of-the-art object detection algorithms, including Mask R-CNN. It is written in Python and powered by the Caffe2 deep learning framework.
|
Psychlab from DeepMind
Psychlab allows you to apply methods from fields like cognitive psychology to study behaviors of artificial agents in a controlled environment. You can read the research paper here.
|
PyTorch implementation of Trust Region Policy Optimization
A PyTorch implementation of Trust Region Policy Optimization (TRPO). In contrast to another implementation of TRPO in PyTorch, this implementation uses exact Hessian-vector product instead of finite differences approximation.
|
|
[1801.07883] Deep Learning for Sentiment Analysis : A Survey
Deep learning has emerged as a powerful machine learning technique that learns multiple layers of representations or features of the data and produces state-of-the-art prediction results. Along with the success of deep learning in many other application domains, deep learning is also popularly used in sentiment analysis in recent years. This paper first gives an overview of deep learning and then provides a comprehensive survey of its current applications in sentiment analysis.
|
[1801.08116] Psychlab: A Psychology Laboratory for Deep Reinforcement Learning Agents
Psychlab is a simulated psychology laboratory inside the first-person 3D game world of DeepMind Lab. Psychlab enables implementations of classical laboratory psychological experiments so that they work with both human and artificial agents. It has a simple and flexible API that enables users to easily create their own tasks.
|
[1701.08230] Algorithmic decision making and the cost of fairness
The authors formulate algorithmic fairness as constrained optimization: the objective is to maximize public safety while satisfying formal fairness constraints designed to reduce racial disparities. The authors focus on algorithms for pretrial release decisions, but the principles apply to other domains, and also to human decision makers carrying out structured decision rules.
|
[1709.04326] Learning with Opponent-Learning Awareness
Opponent-Learning Awareness (LOLA) is a method that reasons about the anticipated learning of the other agents. The LOLA learning rule includes an additional term that accounts for the impact of the agent’s policy on the anticipated parameter update of the other agents. Preliminary results show that the encounter of two LOLA agents leads to the emergence of tit-for-tat and therefore cooperation in the iterated prisoners’ dilemma (IPD), while independent learning does not.
|
Did you enjoy this issue?
|
|
|
|
If you don't want these updates anymore, please unsubscribe here.
If you were forwarded this newsletter and you like it, you can subscribe here.
|
|
|