arXiv subjects

Michael Zhao

Publications and source records attributed to Michael Zhao.

9 recordsLinked to original sources

arXiv2026

Autonomous route to the strange attractor: the many life stages of the chaotic bubblewheel

A body floating atop a supersaturated fluid may accumulate bubbles along its underbelly, which can render the body rotationally unstable. But rotation can strip the surface of these bubbles when they make contact with the air above. To explore this coupling we perform experiments using cylinders floating on carbonated water. The cylinders exhibit an array of distinct dynamics, including constant rolling, periodic and aperiodic oscillation, chaos, and intermittent capsizing, which can be tuned by body size, mass distribution, and gas concentration. A continuum model for the bubble density dynamics, and Galerkin projection, tie the experiment to the generalized Lorenz system and its famed chaotic attractor. Even an extremely small center-of-mass offset can have a qualitative impact on the dynamics, and an offset of as little as a few percent can fully stabilize the system. The experiment represents a highly accessible physical realization of the Lorenz system, which, owing to continuous gas loss from the fluid to the air above, sweeps slowly without intervention across a classical bifurcation diagram in time.

physics.flu-dyn↗

arXiv2026

Experimental Assortments for Choice Estimation and Nest Identification

What assortments (subsets of items) should be offered, to collect data for estimating a choice model over $n$ total items? We propose a structured, non-adaptive experiment design requiring only $O(\log n)$ distinct assortments, each offered repeatedly, that consistently outperforms randomized and other heuristic designs across an extensive numerical benchmark that estimates multiple different choice models under a variety of (possibly mis-specified) ground truths. We then focus on Nested Logit choice models, which cluster items into "nests" of close substitutes. Whereas existing Nested Logit estimation procedures assume the nests to be known and fixed, we present a new algorithm to identify nests based on collected data, which when used in conjunction with our experiment design, guarantees correct identification of nests under any Nested Logit ground truth. Our experiment design was deployed to collect data from over 70 million users at Dream11, an Indian fantasy sports platform that offers different types of betting contests, with rich substitution patterns between them. We identify nests based on the collected data, which lead to better out-of-sample choice prediction than ex-ante clustering from contest features. Our identified nests are ex-post justifiable to Dream11 management.

stat.ME↗

arXiv2025

Advertising Spillovers in Mobile Apps: Evidence from Ad Shutoffs and Store Rankings

Using advertising campaign data from a large US-based mobile game developer, the authors study a global advertising shutoff in the context of mobile app install ads. Contrary to prior studies in search advertising, which reveal a major over-attribution problem where paid advertising takes credit for organic traffic that would have occurred otherwise, this study shows the opposite: paid ads generate positive spillovers to organic installs. Event study analysis shows that the shutoff decreased organic installs by 20-30%. Fixed-effects panel models estimated on longer-term data find that every $100 spent is associated with 32 paid installs and 2.2 organic installs, highly consistent with the event study estimates. Further analysis strongly suggests that this positive paid-to-organic spillover operates through a ranking mechanism: paid installs boost app store category rankings, thereby increasing organic visibility. Combining campaign and ranking data, the authors find that (1) ad spend has a statistically and economically significant relationship with store rankings; and (2) the relationship between organic installs and ad spend disappears once these rankings are factored in, indicating that they absorb the relationship. These findings demonstrate that mobile app install ads are more effective than paid install metrics alone indicate, implying that developers may systematically underinvest in marketing.

econ.GN↗

arXiv2024

QuSpin Zero-Field Magnetometer Characterization for the TUCAN Experiment

The TUCAN nEDM experiment characterizes the QuSpin Zero-Field Magnetometer (QZFM) to accurately map residual fields within a large magnetically shielded room. The magnetometer's intrinsic offset was measured to be within 3 nT and stable over a period of one year. The response was shown to be within 2 percent of linearity in the zero-field regime, up to 2 nTpp, and then follows a smooth dispersion curve. Crosstalk effects induced by multisensor operation were determined to have a small effect, and inconsequential with a separation above 6 cm. These results enable the QZFM for accurate measurement of DC fields, increase the operational range of QZFM by a factor of more than an order of magnitude, and allow for higher efficiency and flexibility by green-lighting simultaneous operation of multiple QZFMs.

physics.ins-det↗

arXiv2023

Evaluating the Surrogate Index as a Decision-Making Tool Using 200 A/B Tests at Netflix

Surrogate index approaches have recently become a popular method of estimating longer-term impact from shorter-term outcomes. In this paper, we leverage 1098 test arms from 200 A/B tests at Netflix to empirically investigate to what degree would decisions made using a surrogate index utilizing 14 days of data would align with those made using direct measurement of day 63 treatment effects. Focusing specifically on linear "auto-surrogate" models that utilize the shorter-term observations of the long-term outcome of interest, we find that the statistical inferences that we would draw from using the surrogate index are ~95% consistent with those from directly measuring the long-term treatment effect. Moreover, when we restrict ourselves to the set of tests that would be "launched" (i.e. positive and statistically significant) based on the 63-day directly measured treatment effects, we find that relying instead on the surrogate index achieves 79% and 65% recall.

stat.AP↗

arXiv2023

Long-Term Causal Inference with Imperfect Surrogates using Many Weak Experiments, Proxies, and Cross-Fold Moments

Inferring causal effects on long-term outcomes using short-term surrogates is crucial to rapid innovation. However, even when treatments are randomized and surrogates fully mediate their effect on outcomes, it's possible that we get the direction of causal effects wrong due to confounding between surrogates and outcomes -- a situation famously known as the surrogate paradox. The availability of many historical experiments offer the opportunity to instrument for the surrogate and bypass this confounding. However, even as the number of experiments grows, two-stage least squares has non-vanishing bias if each experiment has a bounded size, and this bias is exacerbated when most experiments barely move metrics, as occurs in practice. We show how to eliminate this bias using cross-fold procedures, JIVE being one example, and construct valid confidence intervals for the long-term effect in new experiments where long-term outcome has not yet been observed. Our methodology further allows to proxy for effects not perfectly mediated by the surrogates, allowing us to handle both confounding and effect leakage as violations of standard statistical surrogacy conditions.

stat.ME↗

arXiv2018

Sentiment Predictability for Stocks

In this work, we present our findings and experiments for stock-market prediction using various textual sentiment analysis tools, such as mood analysis and event extraction, as well as prediction models, such as LSTMs and specific convolutional architectures.

cs.CL↗

arXiv2017

Binary hermitian forms and optimal embeddings

Fix a quadratic order over the ring of integers. An embedding of the quadratic order into a quaternionic order naturally gives an integral binary hermitian form over the quadratic order. We show that, in certain cases, this correspondence is a discriminant preserving bijection between the isomorphism classes of embeddings and integral binary hermitian forms.

math.NT↗

arXiv2017

Criteria for guaranteed breakdown in two-phase inhomogeneous bodies

Lower bounds are obtained on the maximum field strength in one or both phases in a body containing two-phases. These bounds only incorporate boundary data that can be obtained from measurements at the surface of the body, and thus may be useful for determining if breakdown has necessarily occurred in one of the phases, or that some other nonlinearities have occurred. It is assumed the response of the phases is linear up to the point of electric, dielectric, or elastic breakdown, or up to the point of the onset of nonlinearities. These bounds are calculated for conductivity, with one or two sets of boundary conditions, for complex conductivity (as appropriate at fixed frequency when the wavelength is much larger than the body, i.e., for quasistatics), and for two-dimensional elasticity. Sometimes the bounds are optimal when the field is constant in one of the phases, and using the algorithm of Kang, Kim, and Milton (2012) a wide variety of inclusion shapes having this property, for appropriately chosen bodies and appropriate boundary conditions, are numerically constructed. Such inclusions are known as $E_Ω$-inclusions.

math.AP↗