Searcharxiv⌕ Search

arXiv subjects

Yu Liu

Publications and source records attributed to Yu Liu.

At least 739 records · Page 41Linked to original sources
arXiv2023

TaskMatrix.AI: Completing Tasks by Connecting Foundation Models with Millions of APIs

Artificial Intelligence (AI) has made incredible progress recently. On the one hand, advanced foundation models like ChatGPT can offer powerful conversation, in-context learning and code generation abilities on a broad range of open-domain tasks. They can also generate high-level solution outlines for domain-specific tasks based on the common sense knowledge they have acquired. However, they still face difficulties with some specialized tasks because they lack enough domain-specific data during pre-training or they often have errors in their neural network computations on those tasks that need accurate executions. On the other hand, there are also many existing models and systems (symbolic-based or neural-based) that can do some domain-specific tasks very well. However, due to the different implementation or working mechanisms, they are not easily accessible or compatible with foundation models. Therefore, there is a clear and pressing need for a mechanism that can leverage foundation models to propose task solution outlines and then automatically match some of the sub-tasks in the outlines to the off-the-shelf models and systems with special functionalities to complete them. Inspired by this, we introduce TaskMatrix.AI as a new AI ecosystem that connects foundation models with millions of APIs for task completion. Unlike most previous work that aimed to improve a single AI model, TaskMatrix.AI focuses more on using existing foundation models (as a brain-like central system) and APIs of other AI models and systems (as sub-task solvers) to achieve diversified tasks in both digital and physical domains. As a position paper, we will present our vision of how to build such an ecosystem, explain each key component, and use study cases to illustrate both the feasibility of this vision and the main challenges we need to address next.

cs.AI↗
arXiv2023

Visualizing the atomic-scale origin of metallic behavior in Kondo insulators

A Kondo lattice is often electrically insulating at low temperatures. However, several recent experiments have detected signatures of bulk metallicity within this Kondo insulating phase. Here we visualize the real-space charge landscape within a Kondo lattice with atomic resolution using a scanning tunneling microscope. We discover nanometer-scale puddles of metallic conduction electrons centered around uranium-site substitutions in the heavy-fermion compound URu$_2$Si$_2$, and around samarium-site defects in the topological Kondo insulator SmB$_6$. These defects disturb the Kondo screening cloud, leaving behind a fingerprint of the metallic parent state. Our results suggest that the mysterious 3D quantum oscillations measured in SmB$_6$ could arise from these Kondo-lattice defects, although we cannot rule out other explanations. Our imaging technique could enable the development of atomic-scale charge sensors using heavy-fermion probes.

cond-mat.str-el↗
arXiv2023

End-to-End Personalized Next Location Recommendation via Contrastive User Preference Modeling

Predicting the next location is a highly valuable and common need in many location-based services such as destination prediction and route planning. The goal of next location recommendation is to predict the next point-of-interest a user might go to based on the user's historical trajectory. Most existing models learn mobility patterns merely from users' historical check-in sequences while overlooking the significance of user preference modeling. In this work, a novel Point-of-Interest Transformer (POIFormer) with contrastive user preference modeling is developed for end-to-end next location recommendation. This model consists of three major modules: history encoder, query generator, and preference decoder. History encoder is designed to model mobility patterns from historical check-in sequences, while query generator explicitly learns user preferences to generate user-specific intention queries. Finally, preference decoder combines the intention queries and historical information to predict the user's next location. Extensive comparisons with representative schemes and ablation studies on four real-world datasets demonstrate the effectiveness and superiority of the proposed scheme under various settings.

cs.IR↗
arXiv2023

Deep Learning-based Eye-Tracking Analysis for Diagnosis of Alzheimer's Disease Using 3D Comprehensive Visual Stimuli

Alzheimer's Disease (AD) causes a continuous decline in memory, thinking, and judgment. Traditional diagnoses are usually based on clinical experience, which is limited by some realistic factors. In this paper, we focus on exploiting deep learning techniques to diagnose AD based on eye-tracking behaviors. Visual attention, as typical eye-tracking behavior, is of great clinical value to detect cognitive abnormalities in AD patients. To better analyze the differences in visual attention between AD patients and normals, we first conduct a 3D comprehensive visual task on a non-invasive eye-tracking system to collect visual attention heatmaps. We then propose a multi-layered comparison convolution neural network (MC-CNN) to distinguish the visual attention differences between AD patients and normals. In MC-CNN, the multi-layered representations of heatmaps are obtained by hierarchical convolution to better encode eye-movement behaviors, which are further integrated into a distance vector to benefit the comprehensive visual task. Extensive experimental results on the collected dataset demonstrate that MC-CNN achieves consistent validity in classifying AD patients and normals with eye-tracking data.

eess.IV↗
arXiv2023

Cones: Concept Neurons in Diffusion Models for Customized Generation

Human brains respond to semantic features of presented stimuli with different neurons. It is then curious whether modern deep neural networks admit a similar behavior pattern. Specifically, this paper finds a small cluster of neurons in a diffusion model corresponding to a particular subject. We call those neurons the concept neurons. They can be identified by statistics of network gradients to a stimulation connected with the given subject. The concept neurons demonstrate magnetic properties in interpreting and manipulating generation results. Shutting them can directly yield the related subject contextualized in different scenes. Concatenating multiple clusters of concept neurons can vividly generate all related concepts in a single image. A few steps of further fine-tuning can enhance the multi-concept capability, which may be the first to manage to generate up to four different subjects in a single image. For large-scale applications, the concept neurons are environmentally friendly as we only need to store a sparse cluster of int index instead of dense float32 values of the parameters, which reduces storage consumption by 90\% compared with previous subject-driven generation methods. Extensive qualitative and quantitative studies on diverse scenarios show the superiority of our method in interpreting and manipulating diffusion models.

cs.CV↗
arXiv2023

Presentation Attack Detection with Advanced CNN Models for Noncontact-based Fingerprint Systems

Touch-based fingerprint biometrics is one of the most popular biometric modalities with applications in several fields. Problems associated with touch-based techniques such as the presence of latent fingerprints and hygiene issues due to many people touching the same surface motivated the community to look for non-contact-based solutions. For the last few years, contactless fingerprint systems are on the rise and in demand because of the ability to turn any device with a camera into a fingerprint reader. Yet, before we can fully utilize the benefit of noncontact-based methods, the biometric community needs to resolve a few concerns such as the resiliency of the system against presentation attacks. One of the major obstacles is the limited publicly available data sets with inadequate spoof and live data. In this publication, we have developed a Presentation attack detection (PAD) dataset of more than 7500 four-finger images and more than 14,000 manually segmented single-fingertip images, and 10,000 synthetic fingertips (deepfakes). The PAD dataset was collected from six different Presentation Attack Instruments (PAI) of three different difficulty levels according to FIDO protocols, with five different types of PAI materials, and different smartphone cameras with manual focusing. We have utilized DenseNet-121 and NasNetMobile models and our proposed dataset to develop PAD algorithms and achieved PAD accuracy of Attack presentation classification error rate (APCER) 0.14\% and Bonafide presentation classification error rate (BPCER) 0.18\%. We have also reported the test results of the models against unseen spoof types to replicate uncertain real-world testing scenarios.

cs.CV↗
arXiv2023

Comparative study of first-principles approaches for effective Coulomb interaction strength $U_{\rm eff}$ between localized $f$-electrons: lanthanide metals as an example

As correlation strength has a key influence on the simulation of strongly correlated materials, many approaches have been proposed to obtain the parameter using first-principles calculations. However, the comparison of the different Coulomb strengths obtained using these approaches and an investigation of the mechanisms behind them are still needed. Taking lanthanide metals as an example, we research the factors that affect the effective Coulomb interaction strength, $U_{\rm eff}$, by local screened Coulomb correction (LSCC), linear response (LR) and constrained random-phase approximation (cRPA) in VASP. The $U^{\rm LSCC}_{\rm eff}$ value increases from 4.75 eV to 7.78 eV, $U^{\rm LR}_{\rm eff}$ is almost stable at about 6.0 eV (except for Eu, Er and Lu), and $U^{\rm cRPA}_{\rm eff}$ shows a two-stage decreasing trend in both light and heavy lanthanides. To investigate these differences, we established a scheme to analyze coexistence and competition between the orbital localization and the screening effect. We find that LSCC and cRPA are dominated by the orbital localization and the screening effect, respectively, whereas LR shows a balance of the competition between the two factors. Additionally, the performance of these approaches is influenced by different starting points from PBE and PBE+$U$, especially for cRPA. Our results provide useful knowledge for understanding the $U_{\rm eff}$ of lanthanide materials, and similar analyses can also be used in the research of other correlation strength simulation approaches.

cond-mat.mtrl-sci↗
arXiv2023

SLOTH: Structured Learning and Task-based Optimization for Time Series Forecasting on Hierarchies

Multivariate time series forecasting with hierarchical structure is widely used in real-world applications, e.g., sales predictions for the geographical hierarchy formed by cities, states, and countries. The hierarchical time series (HTS) forecasting includes two sub-tasks, i.e., forecasting and reconciliation. In the previous works, hierarchical information is only integrated in the reconciliation step to maintain coherency, but not in forecasting step for accuracy improvement. In this paper, we propose two novel tree-based feature integration mechanisms, i.e., top-down convolution and bottom-up attention to leverage the information of the hierarchical structure to improve the forecasting performance. Moreover, unlike most previous reconciliation methods which either rely on strong assumptions or focus on coherent constraints only,we utilize deep neural optimization networks, which not only achieve coherency without any assumptions, but also allow more flexible and realistic constraints to achieve task-based targets, e.g., lower under-estimation penalty and meaningful decision-making loss to facilitate the subsequent downstream tasks. Experiments on real-world datasets demonstrate that our tree-based feature integration mechanism achieves superior performances on hierarchical forecasting tasks compared to the state-of-the-art methods, and our neural optimization networks can be applied to real-world tasks effectively without any additional effort under coherence and task-based constraints

cs.LG↗
arXiv2023

Knowledge-infused Contrastive Learning for Urban Imagery-based Socioeconomic Prediction

Monitoring sustainable development goals requires accurate and timely socioeconomic statistics, while ubiquitous and frequently-updated urban imagery in web like satellite/street view images has emerged as an important source for socioeconomic prediction. Especially, recent studies turn to self-supervised contrastive learning with manually designed similarity metrics for urban imagery representation learning and further socioeconomic prediction, which however suffers from effectiveness and robustness issues. To address such issues, in this paper, we propose a Knowledge-infused Contrastive Learning (KnowCL) model for urban imagery-based socioeconomic prediction. Specifically, we firstly introduce knowledge graph (KG) to effectively model the urban knowledge in spatiality, mobility, etc., and then build neural network based encoders to learn representations of an urban image in associated semantic and visual spaces, respectively. Finally, we design a cross-modality based contrastive learning framework with a novel image-KG contrastive loss, which maximizes the mutual information between semantic and visual representations for knowledge infusion. Extensive experiments of applying the learnt visual representations for socioeconomic prediction on three datasets demonstrate the superior performance of KnowCL with over 30\% improvements on $R^2$ compared with baselines. Especially, our proposed KnowCL model can apply to both satellite and street imagery with both effectiveness and transferability achieved, which provides insights into urban imagery-based socioeconomic prediction.

cs.CV↗
arXiv2023

Deep Learning for Video-Text Retrieval: a Review

Video-Text Retrieval (VTR) aims to search for the most relevant video related to the semantics in a given sentence, and vice versa. In general, this retrieval task is composed of four successive steps: video and textual feature representation extraction, feature embedding and matching, and objective functions. In the last, a list of samples retrieved from the dataset is ranked based on their matching similarities to the query. In recent years, significant and flourishing progress has been achieved by deep learning techniques, however, VTR is still a challenging task due to the problems like how to learn an efficient spatial-temporal video feature and how to narrow the cross-modal gap. In this survey, we review and summarize over 100 research papers related to VTR, demonstrate state-of-the-art performance on several commonly benchmarked datasets, and discuss potential challenges and directions, with the expectation to provide some insights for researchers in the field of video-text retrieval.

cs.CV↗
arXiv2023

Composer: Creative and Controllable Image Synthesis with Composable Conditions

Recent large-scale generative models learned on big data are capable of synthesizing incredible images yet suffer from limited controllability. This work offers a new generation paradigm that allows flexible control of the output image, such as spatial layout and palette, while maintaining the synthesis quality and model creativity. With compositionality as the core idea, we first decompose an image into representative factors, and then train a diffusion model with all these factors as the conditions to recompose the input. At the inference stage, the rich intermediate representations work as composable elements, leading to a huge design space (i.e., exponentially proportional to the number of decomposed factors) for customizable content creation. It is noteworthy that our approach, which we call Composer, supports various levels of conditions, such as text description as the global information, depth map and sketch as the local guidance, color histogram for low-level details, etc. Besides improving controllability, we confirm that Composer serves as a general framework and facilitates a wide range of classical generative tasks without retraining. Code and models will be made available.

cs.CV↗
arXiv2023

Improving Object-centric Learning with Query Optimization

The ability to decompose complex natural scenes into meaningful object-centric abstractions lies at the core of human perception and reasoning. In the recent culmination of unsupervised object-centric learning, the Slot-Attention module has played an important role with its simple yet effective design and fostered many powerful variants. These methods, however, have been exceedingly difficult to train without supervision and are ambiguous in the notion of object, especially for complex natural scenes. In this paper, we propose to address these issues by investigating the potential of learnable queries as initializations for Slot-Attention learning, uniting it with efforts from existing attempts on improving Slot-Attention learning with bi-level optimization. With simple code adjustments on Slot-Attention, our model, Bi-level Optimized Query Slot Attention, achieves state-of-the-art results on 3 challenging synthetic and 7 complex real-world datasets in unsupervised image segmentation and reconstruction, outperforming previous baselines by a large margin. We provide thorough ablative studies to validate the necessity and effectiveness of our design. Additionally, our model exhibits great potential for concept binding and zero-shot learning. Our work is made publicly available at https://bo-qsa.github.io

cs.CV↗
arXiv2023

Channel Tracking for RIS-aided mmWave Communications Under High Mobility Scenarios

The emerging reconfigurable intelligent surface (RIS) technology is promising for applications in the millimeter wave (mmWave) communication systems to effectively compensate for propagation loss or tackle the blockage issue. Considering the high mobility of users in realistic scenarios, it is essential to adjust the phase shifts in real time to align the beam towards the mobile users, which requires to frequently estimate the channel state information. Hence, it is imperative to design efficient channel tracking schemes to avoid the complex channel estimation procedure. In this paper, we develop a novel channel tracking scheme with two advantages over conventional schemes. First, our tracking scheme is based on the cascaded angles at the RIS instead of the accurate angle values, which is more practical. Second, it can be employed under a more general setting where the noise can be non-Gaussian. Simulation results show the high tracking accuracy of our proposed scheme, and validate the superiority to the existing EKF-based tracking scheme.

eess.SP↗
arXiv2023

Pressure-induced magnetic properties of quasi-2D Cr2Si2Te6 and Mn3Si2Te6

Recently, the pressure has been used as external stimuli to induce structural and magnetic phase transitions in many layered quantum materials whose layers are linked by van der Waals forces. Such materials with weakly held layers allow relatively easy manipulation of the superexchange mechanism and lead to novel magnetic behavior. Using the hydrostatic pressure as a disorderless means to manipulate the interlayer coupling, we applied pressure on two quasi-2D sister compounds, namely, Cr2Si2Te6 (CST) and Mn3Si2Te6 (MST), up to ~1 GPa. Magnetic property measurements with the application of pressure revealed that the ferromagnetic transition temperature decreases in CST while the opposite trend occurs for the ferrimagnet MST. In MST, the magnetization decreases with the increase in the pressure, and such trend is not clearly noticed in CST, within the pressure range studied. Theoretical calculations showed the overall pressure effect on layer separation, bond angle, and exchange coupling, strongly influencing the change in subsequent magnetic characteristics. Exchange coupling in Mn3Si2Te6 is strongly frustrated and the first nearest neighbor interaction is the most dominant of the components with the strongest pressure dependence. Whereas, in Cr2Si2Te6, the exchange coupling parameters exhibit very little dependence on the pressure. This combined experimental and theoretical work has the potential to expand to other relevant quantum materials.

cond-mat.mtrl-sci↗
arXiv2023

Centralized Cooperative Exploration Policy for Continuous Control Tasks

The deep reinforcement learning (DRL) algorithm works brilliantly on solving various complex control tasks. This phenomenal success can be partly attributed to DRL encouraging intelligent agents to sufficiently explore the environment and collect diverse experiences during the agent training process. Therefore, exploration plays a significant role in accessing an optimal policy for DRL. Despite recent works making great progress in continuous control tasks, exploration in these tasks has remained insufficiently investigated. To explicitly encourage exploration in continuous control tasks, we propose CCEP (Centralized Cooperative Exploration Policy), which utilizes underestimation and overestimation of value functions to maintain the capacity of exploration. CCEP first keeps two value functions initialized with different parameters, and generates diverse policies with multiple exploration styles from a pair of value functions. In addition, a centralized policy framework ensures that CCEP achieves message delivery between multiple policies, furthermore contributing to exploring the environment cooperatively. Extensive experimental results demonstrate that CCEP achieves higher exploration capacity. Empirical analysis shows diverse exploration styles in the learned policies by CCEP, reaping benefits in more exploration regions. And this exploration capacity of CCEP ensures it outperforms the current state-of-the-art methods across multiple continuous control tasks shown in experiments.

cs.LG↗
arXiv2023

Semantic Encoder Guided Generative Adversarial Face Ultra-Resolution Network

Face super-resolution is a domain-specific image super-resolution, which aims to generate High-Resolution (HR) face images from their Low-Resolution (LR) counterparts. In this paper, we propose a novel face super-resolution method, namely Semantic Encoder guided Generative Adversarial Face Ultra-Resolution Network (SEGA-FURN) to ultra-resolve an unaligned tiny LR face image to its HR counterpart with multiple ultra-upscaling factors (e.g., 4x and 8x). The proposed network is composed of a novel semantic encoder that has the ability to capture the embedded semantics to guide adversarial learning and a novel generator that uses a hierarchical architecture named Residual in Internal Dense Block (RIDB). Moreover, we propose a joint discriminator which discriminates both image data and embedded semantics. The joint discriminator learns the joint probability distribution of the image space and latent space. We also use a Relativistic average Least Squares loss (RaLS) as the adversarial loss to alleviate the gradient vanishing problem and enhance the stability of the training procedure. Extensive experiments on large face datasets have proved that the proposed method can achieve superior super-resolution results and significantly outperform other state-of-the-art methods in both qualitative and quantitative comparisons.

cs.CV↗
arXiv2023

Real-World Image Super Resolution via Unsupervised Bi-directional Cycle Domain Transfer Learning based Generative Adversarial Network

Deep Convolutional Neural Networks (DCNNs) have exhibited impressive performance on image super-resolution tasks. However, these deep learning-based super-resolution methods perform poorly in real-world super-resolution tasks, where the paired high-resolution and low-resolution images are unavailable and the low-resolution images are degraded by complicated and unknown kernels. To break these limitations, we propose the Unsupervised Bi-directional Cycle Domain Transfer Learning-based Generative Adversarial Network (UBCDTL-GAN), which consists of an Unsupervised Bi-directional Cycle Domain Transfer Network (UBCDTN) and the Semantic Encoder guided Super Resolution Network (SESRN). First, the UBCDTN is able to produce an approximated real-like LR image through transferring the LR image from an artificially degraded domain to the real-world LR image domain. Second, the SESRN has the ability to super-resolve the approximated real-like LR image to a photo-realistic HR image. Extensive experiments on unpaired real-world image benchmark datasets demonstrate that the proposed method achieves superior performance compared to state-of-the-art methods.

cs.CV↗
arXiv2023

Epitaxially Driven Phase Selectivity of Sn in Hybrid Quantum Nanowires

Hybrid semiconductor/superconductor nanowires constitute a pervasive platform for studying gate-tunable superconductivity and the emergence of topological behavior. Their low-dimensionality and crystal structure flexibility facilitate novel heterostructure growth and efficient material optimization; crucial prerequisites for accurately constructing complex multi-component quantum materials. Here, we present an extensive optimization of Sn growth on InSb, InAsSb and InAs nanowires. We demonstrate how the growth conditions and the crystal structure/symmetry of the semiconductor drive the formation of either semi-metallic $\mathrm{α-Sn}$ or superconducting $\mathrm{β-Sn}$. For InAs nanowires, we obtain phase-pure, superconducting $\mathrm{β-Sn}$ shells. However, for InSb and InAsSb nanowires, an initial epitaxial $\mathrm{α-Sn}$ phase evolves into a polycrystalline shell of coexisting $\mathrmα$ and $\mathrmβ$ phases, where the $β/α$ volume ratio increases with Sn shell thickness. Whether these nanowires exhibit superconductivity or not critically relies on the $\mathrm{β-Sn}$ content. Therefore, this work provides key insights into Sn phase control on a variety of semiconductors, with consequences for the yield of superconducting hybrids suitable for generating topological systems.

cond-mat.mtrl-sci↗