Michael McCourt
YOU?
Author Swipe
Discovering high-performance broadband and broad angle antireflection surfaces by machine learning: erratum Open
In our previously published article [ Optica 7 , 784 ( 2020 ) 10.1364/OPTICA.387938 ], Eqs. (3a), (3b), and (4) were incorrectly stated. This erratum provides the correct equations.
Achieving Diversity in Objective Space for Sample-Efficient Search of Multiobjective Optimization Problems Open
Efficiently solving multi-objective optimization problems for simulation\noptimization of important scientific and engineering applications such as\nmaterials design is becoming an increasingly important research topic. This is\ndue largel…
View article: Bayesian Optimization is Superior to Random Search for Machine Learning Hyperparameter Tuning: Analysis of the Black-Box Optimization Challenge 2020
Bayesian Optimization is Superior to Random Search for Machine Learning Hyperparameter Tuning: Analysis of the Black-Box Optimization Challenge 2020 Open
This paper presents the results and insights from the black-box optimization (BBO) challenge at NeurIPS 2020 which ran from July-October, 2020. The challenge emphasized the importance of evaluating derivative-free optimizers for tuning the…
Processing adjunct control: Evidence on the use of structural information and prediction in reference resolution Open
The comprehension of anaphoric relations may be guided not only by discourse, but also syntactic information. In the literature on online processing, however, the focus has been on audible pronouns and descriptions whose reference is resol…
Discovering high-performance broadband and broad angle antireflection surfaces by machine learning Open
Eliminating light reflection from the top glass sheet in optoelectronic applications is often desirable across a broad range of wavelengths and large variety of angles. In this paper, we report on a combined simulation and experimental stu…
View article: Efficient Rollout Strategies for Bayesian Optimization
Efficient Rollout Strategies for Bayesian Optimization Open
Bayesian optimization (BO) is a class of sample-efficient global optimization methods, where a probabilistic model conditioned on previous observations is used to determine future evaluations via the optimization of an acquisition function…
Discovering Near-Perfect Broadband and Broad Angle Antireflection Surfaces for Optoelectronics by Machine Learning Open
Eliminating light reflection from the top glass sheet in optoelectronic applications is often desirable across a broad range of wavelengths and large variety of angles. In this paper, we report on a combined simulation and experimental stu…
Sampling Humans for Optimizing Preferences in Coloring Artwork Open
Many circumstances of practical importance have performance or success metrics which exist implicitly---in the eye of the beholder, so to speak. Tuning aspects of such problems requires working without defined metrics and only considering …
View article: Bayesian Optimization with Approximate Set Kernels
Bayesian Optimization with Approximate Set Kernels Open
We propose a practical Bayesian optimization method over sets, to minimize a black-box function that takes a set as a single input. Because set inputs are permutation-invariant, traditional Gaussian process-based Bayesian optimization stra…
View article: Practical Bayesian Optimization over Sets
Practical Bayesian Optimization over Sets Open
We propose a practical Bayesian optimization method over sets, to minimize a black-box function that can take a set as a single input. Because set inputs are permutation-invariant and variable-length, traditional Gaussian process-based Bay…
View article: Bayesian Optimization over Sets.
Bayesian Optimization over Sets. Open
We propose a Bayesian optimization method over sets, to minimize a black-box function that can take a set as single input. Because set inputs are permutation-invariant and variable-length, traditional Gaussian process-based Bayesian optimi…
Orchestrate: Infrastructure for Enabling Parallelism during Hyperparameter Optimization Open
Two key factors dominate the development of effective production grade machine learning models. First, it requires a local software implementation and iteration process. Second, it requires distributed infrastructure to efficiently conduct…
A Nonstationary Designer Space-Time Kernel Open
In spatial statistics, kriging models are often designed using a stationary covariance structure; this translation-invariance produces models which have numerous favorable properties. This assumption can be limiting, though, in circumstanc…
Dynamical Properties of Eccentric Nuclear Disks: Stability, Longevity, and Implications for Tidal Disruption Rates in Post-merger Galaxies Open
In some galaxies, the stars orbiting the supermassive black hole take the form of an eccentric nuclear disk, in which every star is on a coherent, apsidally aligned orbit. The most famous example of an eccentric nuclear disk is the double …
The impact of magnetic fields on thermal instability Open
Cold ($T\\sim 10^{4} \\ \\mathrm{K}$) gas is very commonly found in both\ngalactic and cluster halos. There is no clear consensus on its origin. Such gas\ncould be uplifted from the central galaxy by galactic or AGN winds.\nAlternatively, …
Sequential Preference-Based Optimization Open
Many real-world engineering problems rely on human preferences to guide their design and optimization. We present PrefOpt, an open source package to simplify sequential optimization tasks that incorporate human preference feedback. Our app…
Practical Bayesian optimization in the presence of outliers Open
Inference in the presence of outliers is an important field of research as outliers are ubiquitous and may arise across a variety of problems and domains. Bayesian optimization is method that heavily relies on probabilistic inference. This…
Resonant line transfer in a fog: using Lyman-alpha to probe tiny structures in atomic gas Open
\n Motivated by observational and theoretical work that suggest very small-scale (≲ 1 pc) structure in the circumgalactic medium of galaxies and in other environments, we study Lyman-α (Lyα) radiative transfer in an extremely clumpy medium…
A characteristic scale for cold gas Open
We find that clouds of optically-thin, pressure-confined gas are prone to fragmentation as they cool below $\sim10^6$ K. This fragmentation follows the lengthscale $\sim{c}_{\text{s}}\,t_{\text{cool}}$, ultimately reaching very small scale…
Active Preference Learning for Personalized Portfolio Construction Open
In financial asset management, choosing a portfolio requires balancing returns, risk, exposure, liquidity, volatility and other factors. These concerns are difficult to compare explicitly, with many asset managers using an intuitive or imp…
Robust Bayesian Optimization with Student-t Likelihood Open
Bayesian optimization has recently attracted the attention of the automatic machine learning community for its excellent results in hyperparameter tuning. BO is characterized by the sample efficiency with which it can optimize expensive bl…
Simulations of Magnetic Fields in Tidally Disrupted Stars Open
We perform the first magnetohydrodynamical simulations of tidal disruptions of stars by supermassive black holes. We consider stars with both tangled and ordered magnetic fields, for both grazing and deeply disruptive encounters. When the …
The impact of star formation feedback on the circumgalactic medium Open
We use idealized 3D hydrodynamic simulations to study the dynamics and\nthermal structure of the circumgalactic medium (CGM). Our simulations quantify\nthe role of cooling, stellar feedback driven galactic winds and cosmological\ngas accre…
Interactive Preference Learning of Utility Functions for Multi-Objective Optimization Open
Real-world engineering systems are typically compared and contrasted using multiple metrics. For practical machine learning systems, performance tuning is often more nuanced than minimizing a single expected loss objective, and it may be m…
Preemptive Termination of Suggestions during Sequential Kriging Optimization of a Brain Activity Reconstruction Simulation Open
Reconstructing brain activity through electroencephalography requires a boundary value problem (BVP) solver to take a proposed distribution of current dipoles within the brain and compute the resulting electrostatic potential on the scalp.…
Bayesian Optimization for Machine Learning : A Practical Guidebook Open
The engineering of machine learning systems is still a nascent field; relying on a seemingly daunting collection of quickly evolving tools and best practices. It is our hope that this guidebook will serve as a useful resource for machine l…
Using gas clouds to probe the accretion flow around Sgr A*: G2's delayed pericentre passage Open
We study the dynamical evolution of the putative gas clouds G1 and G2 recently discovered in the Galactic Centre. Following earlier studies suggesting that these two clouds are part of a larger gas streamer, we combine their orbits into a …
Evaluation System for a Bayesian Optimization Service Open
Bayesian optimization is an elegant solution to the hyperparameter optimization problem in machine learning. Building a reliable and robust Bayesian optimization service requires careful testing methodology and sound statistical analysis. …