ASAP integrates an LLM agent over a pool of HPO tools and adds system-level optimizations (prefix-stable prompts, speculation parallelism, Self-Tuner) to improve end-to-end wall-clock performance on diverse HPO tasks.
Title resolution pending
5 Pith papers cite this work, alongside 144 external citations. Polarity classification is still indexing.
5
Pith papers citing it
144
external citations · external index
verdicts
UNVERDICTED 5representative citing papers
Sub-surface voids convert a single shock into weaker timed shocks that passively suppress Richtmyer-Meshkov instability growth by over 70% in low-pressure surrogate targets.
ENN surrogate replaces GPs in TuRBO, delivering O(N) scaling for fitting and acquisition and 10-100x faster proposals on large observation counts.
A diversity-aware selection framework builds materials datasets that improve prediction performance on both targeted (up to 25% gain) and untargeted properties (up to 10% gain) compared to random or non-diverse sampling in noisy experimental settings.
Synthetic simulations show noise hurts needle-in-haystack optimization far more than smooth landscapes with local optima, and prior domain knowledge of noise and structure is needed for effective BO in materials research.
citing papers explorer
-
ASAP: Agent-System Co-Design for Wall-Clock-Centered Auto HPO Research for ML Experiments
ASAP integrates an LLM agent over a pool of HPO tools and adds system-level optimizations (prefix-stable prompts, speculation parallelism, Self-Tuner) to improve end-to-end wall-clock performance on diverse HPO tasks.
-
Passive freeze-out of the Richtmyer-Meshkov instability
Sub-surface voids convert a single shock into weaker timed shocks that passively suppress Richtmyer-Meshkov instability growth by over 70% in low-pressure surrogate targets.
-
Taking the GP Out of the Loop
ENN surrogate replaces GPs in TuRBO, delivering O(N) scaling for fitting and acquisition and 10-100x faster proposals on large observation counts.
-
Building informative materials datasets beyond targeted objectives
A diversity-aware selection framework builds materials datasets that improve prediction performance on both targeted (up to 25% gain) and untargeted properties (up to 10% gain) compared to random or non-diverse sampling in noisy experimental settings.
-
Multi-Variable Batch Bayesian Optimization in Materials Research: Synthetic Data Analysis of Noise Sensitivity and Problem Landscape Effects
Synthetic simulations show noise hurts needle-in-haystack optimization far more than smooth landscapes with local optima, and prior domain knowledge of noise and structure is needed for effective BO in materials research.