REVIEW 5 cited by
The Vizier Gaussian Process Bandit Algorithm
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Google Vizier has performed millions of optimizations and accelerated numerous research and production systems at Google, demonstrating the success of Bayesian optimization as a large-scale service. Over multiple years, its algorithm has been improved considerably, through the collective experiences of numerous research efforts and user feedback. In this technical report, we discuss the implementation details and design choices of the current default algorithm provided by Open Source Vizier. Our experiments on standardized benchmarks reveal its robustness and versatility against well-established industry baselines on multiple practical modes.
Forward citations
Cited by 5 Pith papers
-
RoboBallet: Planning for Multi-Robot Reaching with Graph Neural Networks and Reinforcement Learning
A graph-neural-network policy trained with reinforcement learning solves joint task allocation, scheduling, and motion planning for multi-robot reaching, scaling to eight arms and 40 tasks with zero-shot generalization.
-
Performance Prediction for Large Systems via Text-to-Text Regression
A 60M-parameter text-to-text regression model predicts Google Borg cluster efficiency from raw system logs, achieving up to 0.99 rank correlation and 100x lower MSE than tabular baselines.
-
Efficient Heteroscedastic Bayesian Optimization for Risk-Aware AutoRL
Adaptive re-sampling of RAHBO finds reliable RL hyperparameters more sample-efficiently than fixed-replication risk-averse or risk-neutral BO on offline multi-seed datasets.
-
Towards Universal Offline Black-Box Optimization via Learning Language Model Embeddings
A string-based, metadata-guided language model framework for universal offline black-box optimization, with two variants and two embedding regularizations.
-
Rethinking Data Mixture for Large Language Models: A Comprehensive Survey and New Perspectives
A survey organizing LLM data mixture methods into offline and online families, with a fine-grained taxonomy based on optimization frameworks.
Discussion (0). Continue with ORCID to comment.