• Corpus ID: 235187419

Guided Hyperparameter Tuning Through Visualization and Inference

  title={Guided Hyperparameter Tuning Through Visualization and Inference},
  author={Hyekang Joo and Calvin Bao and Ishan Sen and Furong Huang and Leilani Battle},
For deep learning practitioners, hyperparameter tuning for optimizing model performance can be a computationally expensive task. Though visualization can help practitioners relate hyperparameter settings to overall model performance, significant manual inspection is still required to guide the hyperparameter settings in the next batch of experiments. In response, we present a streamlined visualization system enabling deep learning practitioners to more efficiently explore, tune, and optimize… 

Figures and Tables from this paper

VisCUIT: Visual Auditor for Bias in CNN Image Classifier
V IS CUIT is an interactive visualization system that reveals how and why a CNN classifier is biased and helps users discover and characterize the cause of the underperformances by revealing image concepts responsible for activating neurons that contribute to misclassi fications.


HyperTendril: Visual Analytics for User-Driven Hyperparameter Optimization of Deep Neural Networks
HyperTendril is a web-based visual analytics system that supports user-driven hyperparameter tuning processes in a model-agnostic environment that allows users to refine the search spaces and the configuration of the AutoML method based on their own insights from given results.
An Efficient Approach for Assessing Hyperparameter Importance
This paper describes efficient methods that can be used to gain insight into the relationship between hyperparameter settings and performance, and demonstrates that--even in very highdimensional cases--most performance variation is attributable to just a few hyperparameters.
Automatic tuning of hyperparameters using Bayesian optimization
This work uses the CIFAR-10 Dataset and applies the Bayesian hyperparameter optimization algorithm to enhance the performance of the model and clearly obtains optimized values for all hyperparameters which saves time and improves performance.
Practical Bayesian Optimization of Machine Learning Algorithms
This work describes new algorithms that take into account the variable cost of learning algorithm experiments and that can leverage the presence of multiple cores for parallel experimentation and shows that these proposed algorithms improve on previous automatic procedures and can reach or surpass human expert-level optimization for many algorithms.
Algorithms for Hyper-Parameter Optimization
This work contributes novel techniques for making response surface models P(y|x) in which many elements of hyper-parameter assignment (x) are known to be irrelevant given particular values of other elements.
Population Based Training of Neural Networks
Population Based Training is presented, a simple asynchronous optimisation algorithm which effectively utilises a fixed computational budget to jointly optimise a population of models and their hyperparameters to maximise performance.
Interacting with Predictions: Visual Inspection of Black-box Machine Learning Models
The design and implementation of an interactive visual analytics system, Prospector, that provides interactive partial dependence diagnostics and support for localized inspection allows data scientists to understand how and why specific datapoints are predicted as they are.
ModelTracker: Redesigning Performance Analysis Tools for Machine Learning
ModelTracker is presented, an interactive visualization that subsumes information contained in numerous traditional summary statistics and graphs while displaying example-level performance and enabling direct error examination and debugging.
Adam: A Method for Stochastic Optimization
This work introduces Adam, an algorithm for first-order gradient-based optimization of stochastic objective functions, based on adaptive estimates of lower-order moments, and provides a regret bound on the convergence rate that is comparable to the best known results under the online convex optimization framework.
A Generalized Framework for Population Based Training
This work proposes a general, black-box PBT framework that distributes many asynchronous "trials" (a small number of training steps with warm-starting) across a cluster, coordinated by the PBT controller, and shows that the system achieves better accuracy and faster convergence compared to existing methods, given the same computational resource.