VIZ4ML : 머신러닝을 위한 시각화 연구 동향

Better data leads to better models. The power of machine learning comes from its ability to learn patterns from large amounts of data. Understanding your data is critical to building a powerful machine learning system. Facets contains two robust visualizations to aid in understanding and analyzing machine learning datasets.

@InProceedings{wang2018atmseer, title={ATMSeer: Increasing Transparency and Controllability in Automated Machine Learnings}, author={Wang, Qianwen and Ming, Yao and Jin, Zhihua and Shen, Qiaomu and Liu, Dongyu and Smith, Micah J. and Veeramachaneni, Kalyan, and Qu, Huamin}, booktitle = {Conference on Human Factors in Computing Systems (CHI)}, year = {2018} }

About HyperTendril: Visual Analytics for User-Driven Hyperparameter Optimization for Deep Neural Networks

Visualize high dimensional data.

PANENE (Progressive Approximate k-NEarest NEighbors) is a novel algorithm for the k-nearest neighbor (KNN) problem. In contrast to previous algorithms such as Annoy, FLANN, or many others from this benchmark, PANENE is progressive: it can process multiple mini-batches online while keeping each iteration bounded in time.

With the growing success of neural networks, there is a corresponding need to be able to explain their decisions - including building confidence about how they will behave in the real-world, detecting model bias, and for scientific curiosity. In order to do so, we need to both construct deep abstractions and reify (or instantiate) them in rich interfaces .

Understanding how neural networks make predictions remains a fundamental challenge. Existing work on interpreting neural network predictions for images often focuses on explaining predictions for single images or neurons, yet predictions are computed from millions of weights optimized over millions of images-such explanations can easily miss a bigger picture.

A prediction can be explained by assuming that each feature value of the instance is a "player" in a game where the prediction is the payout. Shapley values - a method from coalitional game theory - tells us how to fairly distribute the "payout" among the features.