FlipboardIcon version of the Flipboard logoAre We Really Making Much Progress? A Worrying Analysis of Recent Neural Recommendation Approaches. (arXiv:1907.06902v1arxiv.org - Maurizio Ferrari Dacrema, Paolo Cremonesi, Dietmar Jannach Deep learning techniques have become the method of choice for researchers working on algorithmic aspects of recommender systems. With the strongly …
Data Science Simulating A/B tests offline with counterfactual inferenceabhadury.com While developing machine learning (ML) algorithms in production environments, we usually optimize a function or a loss that has nothing to do with …
A few Recommendations for a Data Scientist who wants to get started in Recommender Systemsbbvadata.com As a Data Scientist, you are expected to be able to build all sort of data products, that may involve simple-yet highly valuable business trends …
The deepest problem with deep learningmedium.com - Gary Marcus Some reflections on an accidental Twitterstorm, the future of AI and deep learning, and what happens when you confuse a schoolbus with a snow plow. On …
NLP's ImageNet moment has arrivedthegradient.pub - Sebastian Ruder Big changes are underway in the world of Natural Language Processing (NLP).The long reign of word vectors as NLP’s core representation technique has …
AdamW and Super-convergence is now the fastest way to train neural netsfast.ai Written: 02 Jul 2018 by Sylvain Gugger and Jeremy Howard Note from Jeremy: Welcome to fast.ai’s first scholar-in-residence, Sylvain Gugger. What better …
Visualizing Matrix Factorization Using Self-Organizing Mapszillow.com One of the core methods used within Zillow’s home recommendation engine is collaborative filtering. There are several ways collaborative filtering …
The Machine Learning Reproducibility Crisispetewarden.com - Pete Warden Gosper Glider GunI was recently chatting to a friend whose startup’s machine learning models were so disorganized it was causing serious problems as …
Black-box optimization — Graduate Descenttimvieira.github.io - by Tim Vieira Black-box optimization algorithms are a fantastic tool that everyone should be aware of. I frequently use black-box optimization algorithms for …
The Building Blocks of Interpretabilitydistill.pub Interpretability techniques are normally studied in isolation. We explore the powerful interfaces that arise when you combine them -- and the rich …
Deep Reinforcement Learning Doesn't Work Yetalexirpan.com June 24, 2018 note: If you want to cite an example from the post, please cite the paper which that example came from. If you want to cite the post as …
How to do machine learning efficientlyhackernoon.com - Radek Osmulski I have just come out of a project where 80% into it I felt I had very little. I invested a lot of time and in the end it was a total fiasco. The math …
How To Create Data Products That Are Magical Using Sequence-to-Sequence Modelstowardsdatascience.com - Hamel Husain A tutorial on how to summarize text and generate features from Github Issues using deep learning with Keras and TensorFlow. The above results are randomly …
AI and Deep Learning in 2017 – A Year in Reviewwildml.com - Denny Britz The year is coming to an end. I did not write nearly as much as I had planned to. But I’m hoping to change that next year, with more tutorials around …
Implementing Poincaré Embeddingsrare-technologies.com - Jayant Jain I have been working on implementing a model called Poincaré embeddings over the last month or so. The model is from an interesting paper by Facebook …
Optimization for Deep Learning Highlights in 2017ruder.io - Sebastian Ruder Table of contents:• Improving Adam• Decoupling weight decay • Fixing the exponential moving average • Tuning the learning rate • Warm restarts• SGD with …
Understanding Hinton’s Capsule Networks. Part I: Intuition.medium.com - Max Pechyonkin CNNs (convolutional neural networks) are awesome. They are one of the reasons deep learning is so popular today. They can do amazing things that …
How to unit test machine learning code.medium.com - Chase Roberts Edit: The popularity of this post has inspired me to write a machine learning test library. Go check it out! Second Edit: The Github user suriyadeepan …
Word embeddings in 2017: Trends and future directionsruder.io - Sebastian Ruder Table of contents:• Subword-level embeddings • OOV handling • Evaluation • Multi-sense embeddings • Beyond words as points • Phrases and multi-word …
Approximate Nearest Neighbours for Recommender Systemsbenfrederickson.com One challenge that recommender systems face is in quickly generating a list of the best recommendations to show for the user. These days many …
Language identificationfasttext.cc We are excited to announce that we are publishing a fast and accurate tool for text-based language identification. It can recognize more than 170 …
Exploring K-Means in Python, C++ and CUDAgoldsborough.me - Peter Goldsborough K-means is a popular clustering algorithm that is not only simple, but also very fast and effective, both as a quick hack to preprocess some data and …
Named Entity Recognition and the Road to Deep Learningnlp.town Not so very long ago, Natural Language Processing looked very different. In sequence labelling tasks such as Named Entity Recognition, Conditional …
datas-frame – Scalable Machine Learning (Part 1)tomaugspurger.github.io - Tom Augspurger This work is supported by Anaconda Inc. and the Data Driven Discovery Initiative from the Moore Foundation. Anaconda is interested in scaling the …
Quote Symbol …Ben Frederickson http://giorgiopatrini.org/posts/2017/09/06/in-search-of-the-missing-signals/
Meet Michelangelo: Uber’s Machine Learning Platformuber.com - Jeremy Hermann & Mike Del Balso Uber Engineering introduces Michelangelo, our machine learning-as-a-service system that enables teams to easily build, deploy, and operate ML …
DrQAgithub.com - facebookresearch This is a PyTorch implementation of the DrQA system described in the ACL 2017 paper Reading Wikipedia to Answer Open-Domain Questions. DrQA is a …
37 Reasons why your Neural Network is not working – Slavslavv.com - Slav Ivanov The network had been training for the last 12 hours. It all looked good: the gradients were flowing and the loss was decreasing. But then came the …
Deep Learning for NLP Best Practicesruder.io - Sebastian Ruder Update July 26, 2017: For additional context, the HackerNews discussion about this post.Table of contents:• Introduction • Best practices• Word …
