Searcharxiv⌕ Search

arXiv subjects

Yuheng Zhang

Publications and source records attributed to Yuheng Zhang.

At least 73 records · Page 4Linked to original sources

Fabrication of Ultra-Low-Loss, Dispersion-Engineered Silicon Nitride Photonic Integrated Circuits via Silicon Hardmask Etching

Silicon nitride (Si$_3$N$_4$) photonic integrated circuits (PICs) have emerged as a versatile platform for a wide range of applications, such as nonlinear optics, narrow-linewidth lasers, and quantum photonics. While thin-film Si$_3$N$_4$ processes have been extensively developed, many nonlinear and quantum optics applications require the use of thick Si$_3$N$_4$ films with engineered dispersion, high mode confinement, and low optical loss. However, high tensile stress in thick Si$_3$N$_4$ films often leads to cracking, making the fabrication challenging to meet these requirements. In this work, we present a robust and reliable fabrication method for ultra-low-loss, dispersion-engineered Si$_3$N$_4$ PICs using amorphous silicon (a-Si) hardmask etching. This approach enables smooth etching of thick Si$_3$N$_4$ waveguides while ensuring long-term storage of crack-free Si$_3$N$_4$ wafers. We achieve intrinsic quality factors ($Q_i$) as high as $25.6 \times 10^6$, corresponding to a propagation loss of 1.6 dB/m. The introduction of a-Si hardmask etching and novel crack-isolation trenches offers notable advantages, including high etching selectivity, long-term wafer storage, high yield, and full compatibility with existing well-developed silicon-based semiconductor processes. We demonstrate frequency comb generation in the fabricated microring resonators, showcasing the platform's potential for applications in optical communication, nonlinear optics, metrology, and spectroscopy. This stable and efficient fabrication method offers high performance with significantly reduced fabrication complexity, representing a remarkable advancement toward mass production of Si$_3$N$_4$ PICs for a wide spectrum of applications.

physics.optics↗

Spider-Webb: JWST Near Infrared Camera resolved galaxy star formation and nuclear activities in the Spiderweb protocluster at z=2.16

Near-infrared (NIR) emission is less affected by dust than UV and optical emission and is therefore useful for studying the properties of dust-obscured galaxies. Although rest-frame NIR observations of high-redshift galaxies have long been made using space telescopes, their structures were unresolved due to the lack of angular resolution. This letter reports the early results from the analysis of high-resolution Pa$β$ imaging of the Spiderweb protocluster at $z=2.16$ with the JWST Near Infrared Camera. We investigate radial profiles of Pa$β$ lines and rest-frame NIR continua from luminous H$α$-emitting galaxies (HAEs) in the protocluster. Particularly, we compare those of 11 HAEs (N-HAEs) on the star-forming main sequence with those of 8 HAEs (X-HAEs) with X-ray AGNs. Resultant composite Pa$β$ line images of N-HAEs indicate significant star formation in galactic disks. In contrast, X-HAEs are dominated by point source components rather than outer star formation, as inferred from our earlier work based on multi-wavelength SED fitting. Given their higher stellar potentials suggested from their rest-frame NIR images, the different characteristics may be driven by the impact of AGN feedback.

astro-ph.GA↗

JWST/NIRCam Pa$\mathrmβ$ narrow-band imaging reveals ordinary dust extinction for H$\mathrmα$ emitters within the Spiderweb protocluster at z=2.16

We combine JWST/NIRCam and Subaru/MOIRCS dual Pa$\mathrmβ$ + H$\mathrmα$ narrow-band imaging to trace the dust attenuation and the star-formation activities of a sample of 43 H$\mathrmα$ emitters at the core of one of the most massive and best-studied clusters in formation at the cosmic noon: the Spiderweb protocluster at $\mathrm{z=2.16}$. We find that most H$\mathrmα$ emitters display Pa$\mathrmβ$/H$\mathrmα$ ratios compatible with Case B recombination conditions, which translates into nebular extinction values ranging at $\mathrm{A_V\approx0-3}$ magnitudes, and dust corrected $\mathrm{Paβ}$ star formation rates consistent with coeval main sequence field galaxies at fixed stellar mass ($\mathrm{9.4<\log M_*/M_\odot<11.0}$) during this cosmic epoch. Furthermore, we investigate possible environmental impacts on dust extinction across the protocluster large-scale structure and find no correlation between the dustiness of its members and environmental proxies such as phase-space position, clustercentric radius, or local density. These results support the scenario for which dust production within the main galaxy population of this protocluster is driven by secular star formation activities fueled by smooth gas accretion across its large-scale structure. This downplays the role of gravitational interactions in boosting star formation and dust production within the Spiderweb protocluster, in contrast with observations in higher redshift and less evolved protocluster cores.

astro-ph.GA↗

JWST/NIRCam Narrowband Survey of Pa$β$ Emitters in the Spiderweb Protocluster at z=2.16

We report the initial result of our Pa$β$ narrowband imaging on a protocluster with the JWST Near Infrared Camera (NIRCam). As NIRCam enables deep narrowband imaging of rest-frame NIR lines at $z>1$, we target one of the most studied protoclusters, the Spiderweb protocluster at $z=2.16$, in which previous studies have confirmed more than a hundred member galaxies. The NIRCam F405N narrowband filter covers in Pa$β$ line the protocluster redshift given by known protocluster members, allowing the search for new member candidates. The weight-corrected color-magnitude diagram obtained 57 sources showing narrowband excesses, 41 of which satisfy further color selection criteria for limiting the sample to Pa$β$ emitter candidates at $z\sim2.16$, and 24 of them do not have H$α$ emitter counterparts. The Pa$β$ emitter candidates appear to follow the spatial distribution of known protocluster members; however, follow-up spectroscopic confirmation is required. Only 17 out of 58 known H$α$-emitting cluster members are selected as Pa$β$ emitters in the current data, albeit the rest fall out of the narrowband selection owing to their small Pa$β$ equivalent widths. We derive Pa$β$ luminosity function in the Spiderweb protocluster, showing a normalization density of $\log{ϕ^\ast}=-2.53\pm0.15$ at a characteristic Pa$β$ luminosity of $\log{L^\ast}=42.33\pm0.17$. Furthermore, we examine the possibility of detecting faint line emitters visible only in the narrow-band image, but find no promising candidates.

astro-ph.GA↗

On the Curses of Future and History in Future-dependent Value Functions for Off-policy Evaluation

We study off-policy evaluation (OPE) in partially observable environments with complex observations, with the goal of developing estimators whose guarantee avoids exponential dependence on the horizon. While such estimators exist for MDPs and POMDPs can be converted to history-based MDPs, their estimation errors depend on the state-density ratio for MDPs which becomes history ratios after conversion, an exponential object. Recently, Uehara et al. [2022a] proposed future-dependent value functions as a promising framework to address this issue, where the guarantee for memoryless policies depends on the density ratio over the latent state space. However, it also depends on the boundedness of the future-dependent value function and other related quantities, which we show could be exponential-in-length and thus erasing the advantage of the method. In this paper, we discover novel coverage assumptions tailored to the structure of POMDPs, such as outcome coverage and belief coverage, which enable polynomial bounds on the aforementioned quantities. As a side product, our analyses also lead to the discovery of new algorithms with complementary properties.

cs.LG↗

Provably Efficient Interactive-Grounded Learning with Personalized Reward

Interactive-Grounded Learning (IGL) [Xie et al., 2021] is a powerful framework in which a learner aims at maximizing unobservable rewards through interacting with an environment and observing reward-dependent feedback on the taken actions. To deal with personalized rewards that are ubiquitous in applications such as recommendation systems, Maghakian et al. [2022] study a version of IGL with context-dependent feedback, but their algorithm does not come with theoretical guarantees. In this work, we consider the same problem and provide the first provably efficient algorithms with sublinear regret under realizability. Our analysis reveals that the step-function estimator of prior work can deviate uncontrollably due to finite-sample effects. Our solution is a novel Lipschitz reward estimator which underestimates the true reward and enjoys favorable generalization performances. Building on this estimator, we propose two algorithms, one based on explore-then-exploit and the other based on inverse-gap weighting. We apply IGL to learning from image feedback and learning from text feedback, which are reward-free settings that arise in practice. Experimental results showcase the importance of using our Lipschitz reward estimator and the overall effectiveness of our algorithms.

cs.LG↗

Efficient Contextual Bandits with Uninformed Feedback Graphs

Bandits with feedback graphs are powerful online learning models that interpolate between the full information and classic bandit problems, capturing many real-life applications. A recent work by Zhang et al. (2023) studies the contextual version of this problem and proposes an efficient and optimal algorithm via a reduction to online regression. However, their algorithm crucially relies on seeing the feedback graph before making each decision, while in many applications, the feedback graph is uninformed, meaning that it is either only revealed after the learner makes her decision or even never fully revealed at all. This work develops the first contextual algorithm for such uninformed settings, via an efficient reduction to online regression over both the losses and the graphs. Importantly, we show that it is critical to learn the graphs using log loss instead of squared loss to obtain favorable regret guarantees. We also demonstrate the empirical effectiveness of our algorithm on a bidding application using both synthetic and real-world data.

cs.LG↗

The Dust Attenuation Scaling Relation of Star-Forming Galaxies in the EAGLE Simulations

Dust attenuation in star-forming galaxies (SFGs), as parameterized by the infrared excess (IRX $\equiv L_{\rm IR}/L_{\rm UV}$), is found to be tightly correlated with star formation rate (SFR), metallicity and galaxy size, following a universal IRX relation up to $z=3$. This scaling relation can provide a fundamental constraint for theoretical models to reconcile galaxy star formation, chemical enrichment, and structural evolution across cosmic time. We attempt to reproduce the universal IRX relation over $0.1\leq z\leq 2.5$ using the EAGLE hydrodynamical simulations and examine sensitive parameters in determining galaxy dust attenuation. Our findings show that while the predicted universal IRX relation from EAGLE approximately aligns with observations at $z\leq 0.5$, noticeable disparities arise at different stellar masses and higher redshifts. Specifically, we investigate how modifying various galaxy parameters can affect the predicted universal IRX relation in comparison to the observed data. We demonstrate that the simulated gas-phase metallicity is the critical quantity for the shape of the predicted universal IRX relation. We find that the influence of the infrared luminosity and infrared excess is less important while galaxy size has virtually no significant effect. Overall, the EAGLE simulations are not able to replicate some of the observed characteristics between IRX and galaxy parameters of SFGs, emphasizing the need for further investigation and testing for our current state-of-the-art theoretical models.

astro-ph.GA↗

Deep Submillimetre and Radio Observations in the SSA22 Field. II. Sub-millimetre source catalogue and number counts

We present the deepest 850 $μ$m map of the SSA22 field to date, utilizing a combination of new and archival observations taken with SCUBA-2, mounted at the James Clerk Maxwell Telescope (JCMT). The mapped area covers an effective region of approximately 0.34 deg$^2$, achieving a boundary sensitivity of 2 mJy beam$^{-1}$, with the deepest central coverage reaching a depth of $σ_\text{rms}$ $\sim$ 0.79 mJy beam$^{-1}$, the confusion noise is estimated to be $\sim$ 0.43 mJy beam$^{-1}$. A catalogue of 850 $μ$m sources in the SSA22 field is generated, identifying 390 sources with single-to-noise ratios above 3.5, out of which 92 sources exceed 5$σ$. The derived intrinsic number counts at 850 $μ$m are found to be in excellent agreement with published surveys. Interestingly, the SSA22 number counts also exhibit an upturn in the brighter flux region, likely attributed to local emitters or lensing objects within the field. On the scale of $\sim$ 0.3 deg$^2$, the 850 $μ$m number counts are unaffected by cosmic variance and align with the blank field. In the deep region ($σ_\text{rms}$ $\leqslant$ 1 mJy), the counts for fluxes below 8 mJy are consistent with the blank field, and the excess in the brighter regime is not significant. Due to the limited number of very bright sources and the insubstantial cosmic variance in our field, we attribute the fluctuations in the number counts primarily to Poisson noise. The SCUBA-2 850 $μ$m detection in the SSA22 field does not exhibit indications of overdensity.

astro-ph.GA↗

FRAD: Front-Running Attacks Detection on Ethereum using Ternary Classification Model

With the evolution of blockchain technology, the issue of transaction security, particularly on platforms like Ethereum, has become increasingly critical. Front-running attacks, a unique form of security threat, pose significant challenges to the integrity of blockchain transactions. In these attack scenarios, malicious actors monitor other users' transaction activities, then strategically submit their own transactions with higher fees. This ensures their transactions are executed before the monitored transactions are included in the block. The primary objective of this paper is to delve into a comprehensive classification of transactions associated with front-running attacks, which aims to equip developers with specific strategies to counter each type of attack. To achieve this, we introduce a novel detection method named FRAD (Front-Running Attacks Detection on Ethereum using Ternary Classification Model). This method is specifically tailored for transactions within decentralized applications (DApps) on Ethereum, enabling accurate classification of front-running attacks involving transaction displacement, insertion, and suppression. Our experimental validation reveals that the Multilayer Perceptron (MLP) classifier offers the best performance in detecting front-running attacks, achieving an impressive accuracy rate of 84.59% and F1-score of 84.60%.

cs.CR↗

Practical Contextual Bandits with Feedback Graphs

While contextual bandit has a mature theory, effectively leveraging different feedback patterns to enhance the pace of learning remains unclear. Bandits with feedback graphs, which interpolates between the full information and bandit regimes, provides a promising framework to mitigate the statistical complexity of learning. In this paper, we propose and analyze an approach to contextual bandits with feedback graphs based upon reduction to regression. The resulting algorithms are computationally practical and achieve established minimax rates, thereby reducing the statistical complexity in real-world applications.

cs.LG↗

What boost galaxy mergers in two massive galaxy protoclusters at z = 2.24

Characterizing the structural properties of galaxies in high-redshift protoclusters is key to our understanding of the environmental effects on galaxy evolution in the early stages of galaxy and structure formation. In this study, we assess the structural properties of 85 and 87 Halpha emission-line candidates (HAEs) in the densest regions of two massive protoclusters, BOSS1244 and BOSS1542, respectively, using HST H-band imaging data. Our results show a true pair fraction of 22+-5 (33+-6) percent in BOSS1244 (BOSS1542), which yields a merger rate of 0.41+-0.09 (0.52+-0.04) per Gyr for massive HAEs with log (M_*/M_sun) > 10.3. This rate is 1.8 (2.8) times higher than that of the general fields at the same epoch. Our sample of HAEs exhibits half-light radii and Sersic indices that cover a broader range than field star-forming galaxies. Additionally, about 15 percent of the HAEs are as compact as the most massive (log(M_*/M_sun) > 11) spheroid-dominated population. These results suggest that the high galaxy density and cold dynamical state (i.e., velocity dispersion of <400 km/s) are key factors that drive galaxy mergers and promote structural evolution in the two protoclusters. Our findings also indicate that both the local environment (on group scales) and the global environment play essential roles in shaping galaxy morphologies in protoclusters. This is evident in the systematic differences observed in the structural properties of galaxies between BOSS1244 and BOSS1542.

astro-ph.GA↗

Offline Learning in Markov Games with General Function Approximation

We study offline multi-agent reinforcement learning (RL) in Markov games, where the goal is to learn an approximate equilibrium -- such as Nash equilibrium and (Coarse) Correlated Equilibrium -- from an offline dataset pre-collected from the game. Existing works consider relatively restricted tabular or linear models and handle each equilibria separately. In this work, we provide the first framework for sample-efficient offline learning in Markov games under general function approximation, handling all 3 equilibria in a unified manner. By using Bellman-consistent pessimism, we obtain interval estimation for policies' returns, and use both the upper and the lower bounds to obtain a relaxation on the gap of a candidate policy, which becomes our optimization objective. Our results generalize prior works and provide several additional insights. Importantly, we require a data coverage condition that improves over the recently proposed "unilateral concentrability". Our condition allows selective coverage of deviation policies that optimally trade-off between their greediness (as approximate best responses) and coverage, and we show scenarios where this leads to significantly better guarantees. As a new connection, we also show how our algorithmic framework can subsume seemingly different solution concepts designed for the special case of two-player zero-sum games.

cs.LG↗

Improved High-Probability Regret for Adversarial Bandits with Time-Varying Feedback Graphs

We study high-probability regret bounds for adversarial $K$-armed bandits with time-varying feedback graphs over $T$ rounds. For general strongly observable graphs, we develop an algorithm that achieves the optimal regret $\widetilde{\mathcal{O}}((\sum_{t=1}^Tα_t)^{1/2}+\max_{t\in[T]}α_t)$ with high probability, where $α_t$ is the independence number of the feedback graph at round $t$. Compared to the best existing result [Neu, 2015] which only considers graphs with self-loops for all nodes, our result not only holds more generally, but importantly also removes any $\text{poly}(K)$ dependence that can be prohibitively large for applications such as contextual bandits. Furthermore, we also develop the first algorithm that achieves the optimal high-probability regret bound for weakly observable graphs, which even improves the best expected regret bound of [Alon et al., 2015] by removing the $\mathcal{O}(\sqrt{KT})$ term with a refined analysis. Our algorithms are based on the online mirror descent framework, but importantly with an innovative combination of several techniques. Notably, while earlier works use optimistic biased loss estimators for achieving high-probability bounds, we find it important to use a pessimistic one for nodes without self-loop in a strongly observable graph.

cs.LG↗

Improved Algorithms for Neural Active Learning

We improve the theoretical and empirical performance of neural-network(NN)-based active learning algorithms for the non-parametric streaming setting. In particular, we introduce two regret metrics by minimizing the population loss that are more suitable in active learning than the one used in state-of-the-art (SOTA) related work. Then, the proposed algorithm leverages the powerful representation of NNs for both exploitation and exploration, has the query decision-maker tailored for $k$-class classification problems with the performance guarantee, utilizes the full feedback, and updates parameters in a more practical and efficient manner. These careful designs lead to an instance-dependent regret upper bound, roughly improving by a multiplicative factor $O(\log T)$ and removing the curse of input dimensionality. Furthermore, we show that the algorithm can achieve the same performance as the Bayes-optimal classifier in the long run under the hard-margin setting in classification problems. In the end, we use extensive experiments to evaluate the proposed algorithm and SOTA baselines, to show the improved empirical performance.

cs.LG↗

Long-term variation of population exposure to PM2.5 in Eastern China: A perspective from SDG 11.6.2

Air pollution (e.g., PM2.5) has a negative effect on human health. Recently, the population-weighted annual mean PM2.5 concentration (PWAM) has been selected as an indicator 11.6.2 in Sustainable Development Goals (SDGs), for various countries to perfrom a long-term monitoring of population exposure to PM2.5 in cities. However, few studies have employed this indicator for a city-level analysis and also in a long-time series (e.g., for decades). To fill this research gap, this study investigates the long-term (2000-2020) variation of population exposure to PM2.5 in Eastern China (including 318 prefecture-level cities). Three categories of open geospatial data (including high-resolution and long-term PM2.5 and population data, and administrative boundary data of cities) are involved for analysis. We found that: 1) A considerable decrease has been observed for the PWAM during 2014-2020. 2) In 2020, the PWAM is for the first time lower than the interim target-1 (35 μg/m3) defined by the World Health Organization for 214 prefecture-level cities in Eastern China, which accounts for 67% of the total population. The results indicates a considerable improvement of air quality in Eastern China. More important, this study illustrates the feasibility of using open geospatial data to monitor the SDG indicator 11.6.2.

cs.CY↗

Streaming Algorithms with Large Approximation Factors

We initiate a broad study of classical problems in the streaming model with insertions and deletions in the setting where we allow the approximation factor $α$ to be much larger than $1$. Such algorithms can use significantly less memory than the usual setting for which $α= 1+ε$ for an $ε\in (0,1)$. We study large approximations for a number of problems in sketching and streaming and the following are some of our results. For the $\ell_p$ norm/quasinorm $\|x\|_p$ of an $n$-dimensional vector $x$, $0 < p \le 2$, we show that obtaining a $\poly(n)$-approximation requires the same amount of memory as obtaining an $O(1)$-approximation for any $M = n^{Θ(1)}$. For estimating the $\ell_p$ norm, $p > 2$, we show an upper bound of $O(n^{1-2/p} (\log n \allowbreak \log M)/α^{2})$ bits for an $α$-approximation, and give a matching lower bound, for almost the full range of $α\geq 1$ for linear sketches. For the $\ell_2$-heavy hitters problem, we show that the known lower bound of $Ω(k \log n\log M)$ bits for identifying $(1/k)$-heavy hitters holds even if we are allowed to output items that are $1/(αk)$-heavy, for almost the full range of $α$, provided the algorithm succeeds with probability $1-O(1/n)$. We also obtain a lower bound for linear sketches that is tight even for constant probability algorithms. For estimating the number $\ell_0$ of distinct elements, we give an $n^{1/t}$-approximation algorithm using $O(t\log \log M)$ bits of space, as well as a lower bound of $Ω(t)$ bits, both excluding the storage of random bits.

cs.DS↗

A Model-Adaptive Clustering Method for Low-Carbon Energy System Optimization

Intermittent renewable energy resources like wind and solar pose great uncertainty of multiple time scales, from minutes to years, on the design and operation of power systems. Energy system optimization models have been developed to find the least-cost solution to matching the uncertainty with flexibility resources. However, input data that capture such multi-time-scale uncertainty are characterized with a long time horizon and bring great difficulty to solving the optimization model. Here we propose an adaptive clustering method based on the decision variables of optimization model to alleviate the computational complexity, in which the energy system is optimized over selected representative time periods instead of the full time horizon. The proposed clustering method is adaptive to various energy system optimization models or settings, because it extracts features from the optimization models. Results show that the proposed clustering method can significantly lower the error in approximating the solution with the full time horizon, compared to traditional clustering methods.

math.OC↗