Machine Learning for Engineers, NLP.js, Intro to Probabilistic Programming, Facebook Field Guide to ML, Auto-Keras, Exemplar-based Colorization
ArviZ is a Python package for exploratory analysis of Bayesian models. Includes functions for posterior analysis, model checking, comparison and diagnostics.
VisualData – Computer Vision Datasets.
nbdime – diffing and merging of Jupyter Notebooks.
NLP.js – node.js NLP library for building chat bots, with entity extraction, sentiment analysis, automatic language identify, and so more.
Netron – viewer for neural network, deep learning and machine learning models. Supports ONNX, Keras, CoreML, Tensorflow, Caffe, MXNet.
Auto-Keras – library for automated machine learning (AutoML). Provides functions to automatically search for architecture and hyperparameters of deep learning models.
Research Papers and Books
"Cross-Entropy Loss Leads To Poor Margins" – Shows that cross-entropy loss is one of the hidden culprits of adversarial examples. Introduce differential training – novel training paradigm that uses a loss function defined on pairs of each class.
"A Brief Introduction to Machine Learning for Engineers". Free Book by Professor Osvaldo Simeone. [237pp].
"Learning a Shared Shape Space for Multimodal Garment Design". Given an input sketch, network infers both the 2D garment sewing patterns & the draped 3D garment mesh together with the underlying body shape.
"An Introduction to Probabilistic Programming" – Free Book is intended as a graduate-level introduction to probabilistic programming languages and methods for inference in probabilistic programs. [218pp]
"Fast, Better Training Trick — Random Gradient" – Multiplying loss by a random number accelerates convergence and reduces oscillation during optimization on Pascal VOC, Cifar, Cityscapes datasets.
"Recent Advances in Object Detection in the Age of Deep Convolutional Neural Networks" – Reviews recent literature in a comprehensive way, and provides an in-depth view of recent advances, current challenges, problem extensions and datasets. [106pp]
"Neural Approaches to Conversational AI" – Surveys recent neural approaches to conversational AI: question answering agents, task-oriented dialogue agents and chatbots. [85pp]
Deep Exemplar-based Colorization – The first end-to-end deep learning approach to controllable colorisation. Code.
"Recycle-GAN: Unsupervised Video Retargeting" – Demonstrate the advantages of spatiotemporal constraints over spatial constraints for image-to-labels, and labels-to-image.
"Attentive Generative Adversarial Network for Raindrop Removal from A Single Image". Main contribution: Visual attention to both generative and discriminative networks. Code.
Posts, Articles, Tutorials
A (Long) Peek into Reinforcement Learning – Review briefly goes over the field of Reinforcement Learning (RL), from fundamental concepts to classic algorithms. [long-read]
Open Machine Learning Course – Free Course by Open Data Science. Perfectly balanced theory & practice, with each topic followed by an assignment.
Simple guide to Neural Arithmetic Logic Units (NALU): Explanation, Intuition and Code. Cool idea capable of learning a diverse set of tasks including: counting objects on image, translating words to numbers, tracking time.
"Interpretable Machine Learning. A Guide for Making Black Box Models Explainable." – Online Book By Christoph Molnar.
Scipy Lecture Notes: One document to learn numerics, science, and data with Python. Quick introduction to central tools and techniques with increasing level of expertise, from beginner to expert.
Video Lectures and Talks
Facebook Field Guide to Machine Learning video series. Six-part video series shares best real-world practices and provides practical tips about how to apply machine-learning capabilities to real-world problems.
Don’t miss out the next digest, make sure you are subscribed to our newsletter.