arXiv subjects

Xi Chen

Publications and source records attributed to Xi Chen.

At least 739 records · Page 41Linked to original sources

arXiv2022

DP$^2$-VAE: Differentially Private Pre-trained Variational Autoencoders

Modern machine learning systems achieve great success when trained on large datasets. However, these datasets usually contain sensitive information (e.g. medical records, face images), leading to serious privacy concerns. Differentially private generative models (DPGMs) emerge as a solution to circumvent such privacy concerns by generating privatized sensitive data. Similar to other differentially private (DP) learners, the major challenge for DPGM is also how to achieve a subtle balance between utility and privacy. We propose DP$^2$-VAE, a novel training mechanism for variational autoencoders (VAE) with provable DP guarantees and improved utility via \emph{pre-training on private data}. Under the same DP constraints, DP$^2$-VAE minimizes the perturbation noise during training, and hence improves utility. DP$^2$-VAE is very flexible and easily amenable to many other VAE variants. Theoretically, we study the effect of pretraining on private data. Empirically, we conduct extensive experiments on image datasets to illustrate our superiority over baselines under various privacy budgets and evaluation metrics.

cs.LG↗

arXiv2022

SoK: Play-to-Earn Projects

Play-to-earn is one of the prospective categories of decentralized applications. The play-to-earn projects combine blockchain technology with entertaining games and finance, attracting various participants. While huge amounts of capital have been poured into these projects, the new crypto niche is considered controversial, and the traditional gaming industry is hesitant to embrace blockchain technology. In addition, there is little systematic research on these projects. In this paper, we delineate play-to-earn projects in terms of economic & governance models and implementation and analyze how blockchain technology can benefit these projects by providing system robustness, transparency, composability, and decentralized governance. We begin by identifying the participants and characterizing the tokens, which are products of composability. We then summarize the roadmap and governance model to exposit there is a transition from centralized governance to decentralized governance. We also classify the implementation of the play-to-earn projects with different extents of robustness and transparency. Finally, we discuss the security & societal challenges for future research in terms of possible attacks, the economics of tokens, and governance.

cs.CR↗

arXiv2022

Time-Optimal Quantum Driving by Variational Circuit Learning

The simulation of quantum dynamics on a digital quantum computer with parameterized circuits has widespread applications in fundamental and applied physics and chemistry. In this context, using the hybrid quantum-classical algorithm, combining classical optimizers and quantum computers, is a competitive strategy for solving specific problems. We put forward its use for optimal quantum control. We simulate the wave-packet expansion of a trapped quantum particle on a quantum device with a finite number of qubits. We then use circuit learning based on gradient descent to work out the intrinsic connection between the control phase transition and the quantum speed limit imposed by unitary dynamics. We further discuss the robustness of our method against errors and demonstrate the absence of barren plateaus in the circuit. The combination of digital quantum simulation and hybrid circuit learning opens up new prospects for quantum optimal control.

quant-ph↗

arXiv2022

Teacher-Student Architecture for Knowledge Learning: A Survey

Although Deep Neural Networks (DNNs) have shown a strong capacity to solve large-scale problems in many areas, such DNNs with voluminous parameters are hard to be deployed in a real-time system. To tackle this issue, Teacher-Student architectures were first utilized in knowledge distillation, where simple student networks can achieve comparable performance to deep teacher networks. Recently, Teacher-Student architectures have been effectively and widely embraced on various knowledge learning objectives, including knowledge distillation, knowledge expansion, knowledge adaption, and multi-task learning. With the help of Teacher-Student architectures, current studies are able to achieve multiple knowledge-learning objectives through lightweight and effective student networks. Different from the existing knowledge distillation surveys, this survey detailedly discusses Teacher-Student architectures with multiple knowledge learning objectives. In addition, we systematically introduce the knowledge construction and optimization process during the knowledge learning and then analyze various Teacher-Student architectures and effective learning schemes that have been leveraged to learn representative and robust knowledge. This paper also summarizes the latest applications of Teacher-Student architectures based on different purposes (i.e., classification, recognition, and generation). Finally, the potential research directions of knowledge learning are investigated on the Teacher-Student architecture design, the quality of knowledge, and the theoretical studies of regression-based learning, respectively. With this comprehensive survey, both industry practitioners and the academic community can learn insightful guidelines about Teacher-Student architectures on multiple knowledge learning objectives.

cs.LG↗

arXiv2022

Magnetic Resonance Spectroscopy Deep Learning Denoising Using Few In Vivo Data

Magnetic Resonance Spectroscopy (MRS) is a noninvasive tool to reveal metabolic information. One challenge of 1H-MRS is the low Signal-Noise Ratio (SNR). To improve the SNR, a typical approach is to perform Signal Averaging (SA) with M repeated samples. The data acquisition time, however, is increased by M times accordingly, and a complete clinical MRS scan takes approximately 10 minutes at a common setting M=128. Recently, deep learning has been introduced to improve the SNR but most of them use the simulated data as the training set. This may hinder the MRS applications since some potential differences, such as acquisition system imperfections, and physiological and psychologic conditions may exist between the simulated and in vivo data. Here, we proposed a new scheme that purely used the repeated samples of realistic data. A deep learning model, Refusion Long Short-Term Memory (ReLSTM), was designed to learn the mapping from the low SNR time-domain data (24 SA) to the high SNR one (128 SA). Experiments on the in vivo brain spectra of 7 healthy subjects, 2 brain tumor patients and 1 cerebral infarction patient showed that only using 20% repeated samples, the denoised spectra by ReLSTM could provide comparable estimated concentrations of metabolites to 128 SA. Compared with the state-of-the-art low-rank denoising method, the ReLSTM achieved the lower relative error and the Cramér-Rao lower bounds in quantifying some important biomarkers. In summary, ReLSTM can perform high-fidelity denoising of the spectra under fast acquisition (24 SA), which would be valuable to MRS clinical studies.

physics.med-ph↗

arXiv2022

Dissecting Deep Metric Learning Losses for Image-Text Retrieval

Visual-Semantic Embedding (VSE) is a prevalent approach in image-text retrieval by learning a joint embedding space between the image and language modalities where semantic similarities would be preserved. The triplet loss with hard-negative mining has become the de-facto objective for most VSE methods. Inspired by recent progress in deep metric learning (DML) in the image domain which gives rise to new loss functions that outperform triplet loss, in this paper, we revisit the problem of finding better objectives for VSE in image-text matching. Despite some attempts in designing losses based on gradient movement, most DML losses are defined empirically in the embedding space. Instead of directly applying these loss functions which may lead to sub-optimal gradient updates in model parameters, in this paper we present a novel Gradient-based Objective AnaLysis framework, or \textit{GOAL}, to systematically analyze the combinations and reweighting of the gradients in existing DML functions. With the help of this analysis framework, we further propose a new family of objectives in the gradient space exploring different gradient combinations. In the event that the gradients are not integrable to a valid loss function, we implement our proposed objectives such that they would directly operate in the gradient space instead of on the losses in the embedding space. Comprehensive experiments have demonstrated that our novel objectives have consistently improved performance over baselines across different visual/text features and model frameworks. We also showed the generalizability of the GOAL framework by extending it to other models using triplet family losses including vision-language model with heavy cross-modal interactions and have achieved state-of-the-art results on the image-text retrieval tasks on COCO and Flick30K.

cs.CV↗

arXiv2022

Curves on K3 surfaces

We complete the remaining cases of the conjecture predicting existence of infinitely many rational curves on K3 surfaces in characteristic zero, prove almost all cases in positive characteristic and improve the proofs of the previously known cases. To achieve this, we introduce two new techniques in the deformation theory of curves on K3 surfaces. Regeneration, a process opposite to specialisation, which preserves the geometric genus and does not require the class of the curve to extend, and the marked point trick, which allows a controlled degeneration of rational curves to integral ones in certain situations. Combining the two proves existence of integral curves of unbounded degree of any geometric genus g for any projective K3 surface in characteristic zero.

math.AG↗

arXiv2022

MuRAG: Multimodal Retrieval-Augmented Generator for Open Question Answering over Images and Text

While language Models store a massive amount of world knowledge implicitly in their parameters, even very large models often fail to encode information about rare entities and events, while incurring huge computational costs. Recently, retrieval-augmented models, such as REALM, RAG, and RETRO, have incorporated world knowledge into language generation by leveraging an external non-parametric index and have demonstrated impressive performance with constrained model sizes. However, these methods are restricted to retrieving only textual knowledge, neglecting the ubiquitous amount of knowledge in other modalities like images -- much of which contains information not covered by any text. To address this limitation, we propose the first Multimodal Retrieval-Augmented Transformer (MuRAG), which accesses an external non-parametric multimodal memory to augment language generation. MuRAG is pre-trained with a mixture of large-scale image-text and text-only corpora using a joint contrastive and generative loss. We perform experiments on two different datasets that require retrieving and reasoning over both images and text to answer a given query: WebQA, and MultimodalQA. Our results show that MuRAG achieves state-of-the-art accuracy, outperforming existing models by 10-20\% absolute on both datasets and under both distractor and full-wiki settings.

cs.CL↗

arXiv2022

X-GOAL: Multiplex Heterogeneous Graph Prototypical Contrastive Learning

Graphs are powerful representations for relations among objects, which have attracted plenty of attention. A fundamental challenge for graph learning is how to train an effective Graph Neural Network (GNN) encoder without labels, which are expensive and time consuming to obtain. Contrastive Learning (CL) is one of the most popular paradigms to address this challenge, which trains GNNs by discriminating positive and negative node pairs. Despite the success of recent CL methods, there are still two under-explored problems. First, how to reduce the semantic error introduced by random topology based data augmentations. Traditional CL defines positive and negative node pairs via the node-level topological proximity, which is solely based on the graph topology regardless of the semantic information of node attributes, and thus some semantically similar nodes could be wrongly treated as negative pairs. Second, how to effectively model the multiplexity of the real-world graphs, where nodes are connected by various relations and each relation could form a homogeneous graph layer. To solve these problems, we propose a novel multiplex heterogeneous graph prototypical contrastive leaning (X-GOAL) framework to extract node embeddings. X-GOAL is comprised of two components: the GOAL framework, which learns node embeddings for each homogeneous graph layer, and an alignment regularization, which jointly models different layers by aligning layer-specific node embeddings. Specifically, the GOAL framework captures the node-level information by a succinct graph transformation technique, and captures the cluster-level information by pulling nodes within the same semantic cluster closer in the embedding space. The alignment regularization aligns embeddings across layers at both node and cluster levels. We evaluate X-GOAL on various real-world datasets and downstream tasks to demonstrate its effectiveness.

cs.LG↗

arXiv2022

WILD-SCAV: Benchmarking FPS Gaming AI on Unity3D-based Environments

Recent advances in deep reinforcement learning (RL) have demonstrated complex decision-making capabilities in simulation environments such as Arcade Learning Environment, MuJoCo, and ViZDoom. However, they are hardly extensible to more complicated problems, mainly due to the lack of complexity and variations in the environments they are trained and tested on. Furthermore, they are not extensible to an open-world environment to facilitate long-term exploration research. To learn realistic task-solving capabilities, we need to develop an environment with greater diversity and complexity. We developed WILD-SCAV, a powerful and extensible environment based on a 3D open-world FPS (First-Person Shooter) game to bridge the gap. It provides realistic 3D environments of variable complexity, various tasks, and multiple modes of interaction, where agents can learn to perceive 3D environments, navigate and plan, compete and cooperate in a human-like manner. WILD-SCAV also supports different complexities, such as configurable maps with different terrains, building structures and distributions, and multi-agent settings with cooperative and competitive tasks. The experimental results on configurable complexity, multi-tasking, and multi-agent scenarios demonstrate the effectiveness of WILD-SCAV in benchmarking various RL algorithms, as well as it is potential to give rise to intelligent agents with generalized task-solving abilities. The link to our open-sourced code can be found here https://github.com/inspirai/wilderness-scavenger.

cs.LG↗

arXiv2022

Autoregressive Uncertainty Modeling for 3D Bounding Box Prediction

3D bounding boxes are a widespread intermediate representation in many computer vision applications. However, predicting them is a challenging task, largely due to partial observability, which motivates the need for a strong sense of uncertainty. While many recent methods have explored better architectures for consuming sparse and unstructured point cloud data, we hypothesize that there is room for improvement in the modeling of the output distribution and explore how this can be achieved using an autoregressive prediction head. Additionally, we release a simulated dataset, COB-3D, which highlights new types of ambiguity that arise in real-world robotics applications, where 3D bounding box prediction has largely been underexplored. We propose methods for leveraging our autoregressive model to make high confidence predictions and meaningful uncertainty measures, achieving strong results on SUN-RGBD, Scannet, KITTI, and our new dataset.

cs.CV↗

arXiv2022

Efficient broadband frequency conversion via shortcut to adiabaticity

The method of adiabatic frequency conversion, in analogy with the two level atomic system, has been put forward recently and verified experimentally to achieve robust frequency mixing processes such as sum and difference frequency generation. Here we present a comparative study of efficient frequency mixing using various techniques of shortcuts to adiabaticity (STA) such as counter-diabatic driving and invariant-based inverse engineering. We show that, it is possible to perform sum frequency generation by properly designing the poling structure of a periodically poled crystal and the coupling between the input lights and the crystal. The required crystal length for frequency conversion is significantly decreases beyond the adiabatic limit. Our approach significantly improves the robustness of the process against the variation in temperature as well as the signal frequency. By introducing a single parameter control technique with constant coupling and combining with the inverse engineering, perturbation theory and optimal control, we show that the phase mismatch can be further optimized with respect to the fluctuations of input wavelength and crystal temperature that results into a novel experimentally realizable mixing scheme.

quant-ph↗

arXiv2022

Crossmodal-3600: A Massively Multilingual Multimodal Evaluation Dataset

Research in massively multilingual image captioning has been severely hampered by a lack of high-quality evaluation datasets. In this paper we present the Crossmodal-3600 dataset (XM3600 in short), a geographically diverse set of 3600 images annotated with human-generated reference captions in 36 languages. The images were selected from across the world, covering regions where the 36 languages are spoken, and annotated with captions that achieve consistency in terms of style across all languages, while avoiding annotation artifacts due to direct translation. We apply this benchmark to model selection for massively multilingual image captioning models, and show superior correlation results with human evaluations when using XM3600 as golden references for automatic metrics.

cs.CV↗

arXiv2022

Deep Baseline Network for Time Series Modeling and Anomaly Detection

Deep learning has seen increasing applications in time series in recent years. For time series anomaly detection scenarios, such as in finance, Internet of Things, data center operations, etc., time series usually show very flexible baselines depending on various external factors. Anomalies unveil themselves by lying far away from the baseline. However, the detection is not always easy due to some challenges including baseline shifting, lacking of labels, noise interference, real time detection in streaming data, result interpretability, etc. In this paper, we develop a novel deep architecture to properly extract the baseline from time series, namely Deep Baseline Network (DBLN). By using this deep network, we can easily locate the baseline position and then provide reliable and interpretable anomaly detection result. Empirical evaluation on both synthetic and public real-world datasets shows that our purely unsupervised algorithm achieves superior performance compared with state-of-art methods and has good practical applications.

cs.LG↗

arXiv2022

Fast transport of Bose-Einstein condensates in anharmonic traps

We present a method to transport Bose-Einstein condensates (BECs) in anharmonic traps and in the presence of atom-atom interactions in short times without residual excitation. Using a combination of a variational approach and inverse engineering methods, we derive a set of Ermakov-like equations that take into account the coupling between the center of mass motion and the breathing mode. By an appropriate inverse engineering strategy of those equations, we then design the trap trajectory to achieve the desired boundary conditions. Numerical examples for cubic or quartic anharmonicities are provided for fast and high-fidelity transport of BECs. Potential applications are atom interferometry and quantum information processing.

cond-mat.quant-gas↗

arXiv2022

Superscreening and polarization control in confined ferroelectric nematic liquids

The combination of large spontaneous polarization and fluidity makes the newly discovered ferroelectric nematic liquid crystalline phase (NF) responsive to electric fields in ways that have no counterpart in other materials. We probe this sensitive field response by confining a NF fluid in microchannels that connect electrodes through straight and curved paths. We find that by applying electric fields as low as E c.a. 0.5 V/mm, the NF phase orders with its polarization smoothly following the winding paths of the channels even when oriented antiparallel to the line connecting positive to negative electrodes, implying analogous behavior of the electric field. Upon inversion of E, the polar order undergoes a complex multistage switching process dominated by electrostatic interactions. Multistage polarization switching dynamics is also found in numerical simulations of a quasi-2D continuum model of NF liquid crystals in microchannels, which also clarify the conditions under which the electric field is guided by the microchannels. Experiments and theory indicate that all observations are direct consequences of the prompt effective screening of electric field components normal to the channel walls. This electric "superscreening" effect emerges as a distinctive property of the NF phase, capable of inducing conditions in which both the polarization and the electric field are guided by microchannels.

cond-mat.soft↗

arXiv2022

Robust quantum control by smooth quasi-square pulses

Robust time-optimal control is known to feature constant (square) pulses. We analyze fast adiabatic dynamics that preserve robustness by using alternative smooth quasi-square pulses, typically represented by hyper-Gaussian pulses. We here consider the two protocols, robust inverse optimization and time-contracted adiabatic passage, allowing the design of the same pulse shape in both cases. The dynamics and their performance are compared. The superiority of the former protocol is shown.

quant-ph↗

arXiv2022

Shortcuts to Adiabaticity for Fast Qubit Readout in Circuit Quantum Electrodynamics

We propose how to engineer the longitudinal coupling to accelerate the measurement of a qubit longitudinally coupled to a cavity, motivated by the concept of shortcuts to adiabaticity. Different modulations are inversely designed from two methods of inverse engineering and counter-diabatic driving, for achieving larger values of the signal-to-noise ratio (SNR) at nanosecond scale. By comparison, we demonstrate that our protocols outperform the usual periodic modulations on the pointer state separation and SNR. Finally, we show a possible implementation considering state-of-the-art circuit quantum electrodynamics architecture, estimating the minimal time allowed for the measurement process.

quant-ph↗