Searcharxiv⌕ Search

arXiv subjects

Yang Ji

Publications and source records attributed to Yang Ji.

At least 37 records · Page 2Linked to original sources
arXiv2024

Contrastive Conditional Alignment based on Label Shift Calibration for Imbalanced Domain Adaptation

Many existing unsupervised domain adaptation (UDA) methods primarily focus on covariate shift, limiting their effectiveness in imbalanced domain adaptation (IDA) where both covariate shift and label shift coexist. Recent IDA methods have achieved promising results based on self-training using target pseudo labels. However, under the IDA scenarios, the classifier learned in the source domain will exhibit different decision bias from the target domain. It will potentially make target pseudo labels unreliable, and will further lead to error accumulation with incorrect class alignment. Thus, we propose contrastive conditional alignment based on label shift calibration (CCA-LSC) for IDA, to address both covariate shift and label shift. Initially, our contrastive conditional alignment resolve covariate shift to learn representations with domain invariance and class discriminability, which include domain adversarial learning, sample-weighted moving average centroid alignment and discriminative feature alignment. Subsequently, we estimate the probability distribution of the target domain, and calibrate target sample classification predictions based on label shift metrics to encourage labeling pseudo-labels more consistently with the distribution of real target data. Extensive experiments are conducted and demonstrate that our method outperforms existing UDA and IDA methods on benchmarks with both label shift and covariate shift. Our code is available at https://github.com/ysxcj-hub/CCA-LSC.

cs.CV↗
arXiv2024

TaE: Task-aware Expandable Representation for Long Tail Class Incremental Learning

Class-incremental learning is dedicated to the development of deep learning models that are capable of acquiring new knowledge while retaining previously learned information. Most methods focus on balanced data distribution for each task, overlooking real-world long-tailed distributions. Therefore, Long-Tailed Class-Incremental Learning has been introduced, which trains on data where head classes have more samples than tail classes. Existing methods mainly focus on preserving representative samples from previous classes to combat catastrophic forgetting. Recently, dynamic network algorithms freeze old network structures and expand new ones, achieving significant performance. However, with the introduction of the long-tail problem, merely extending Determined blocks can lead to miscalibrated predictions, while expanding the entire backbone results in an explosion of memory size. To address these issues, we introduce a novel Task-aware Expandable (TaE) framework, dynamically allocating and updating task-specific trainable parameters to learn diverse representations from each incremental task while resisting forgetting through the majority of frozen model parameters. To further encourage the class-specific feature representation, we develop a Centroid-Enhanced (CEd) method to guide the update of these task-aware parameters. This approach is designed to adaptively allocate feature space for every class by adjusting the distance between intra- and inter-class features, which can extend to all "training from sketch" algorithms. Extensive experiments demonstrate that TaE achieves state-of-the-art performance.

cs.CV↗
arXiv2024

Extension of a Pattern Recognition Validation Approach for Noisy Boson Sampling

Boson sampling is one of the main quantum computation models to demonstrate the quantum computational advantage. However, this aim may be hard to realize considering two main kinds of noises, which are photon distinguishability and photon loss. Inspired by the Bayesian validation extended to evaluate whether distinguishability is too high to demonstrate this advantage, the pattern recognition validation is extended for boson sampling, considering both distinguishability and loss. Based on clusters constructed with the K means++ method, where parameters are carefully adjusted to optimize the extended validation performances, the distribution of characteristic values is nearly monotonically changed with indistinguishability, especially when photons are close to be indistinguishable. However, this regulation may be suppressed by photon loss. The intrinsic data structure of output events is analyzed through calculating probability distributions and mean 2-norm distances of the sorted outputs. An approximation algorithm is also used to show the data structure changes with noises.

quant-ph↗
arXiv2024

Decoupled Federated Learning on Long-Tailed and Non-IID data with Feature Statistics

Federated learning is designed to enhance data security and privacy, but faces challenges when dealing with heterogeneous data in long-tailed and non-IID distributions. This paper explores an overlooked scenario where tail classes are sparsely distributed over a few clients, causing the models trained with these classes to have a lower probability of being selected during client aggregation, leading to slower convergence rates and poorer model performance. To address this issue, we propose a two-stage Decoupled Federated learning framework using Feature Statistics (DFL-FS). In the first stage, the server estimates the client's class coverage distributions through masked local feature statistics clustering to select models for aggregation to accelerate convergence and enhance feature learning without privacy leakage. In the second stage, DFL-FS employs federated feature regeneration based on global feature statistics and utilizes resampling and weighted covariance to calibrate the global classifier to enhance the model's adaptability to long-tailed data distributions. We conducted experiments on CIFAR10-LT and CIFAR100-LT datasets with various long-tailed rates. The results demonstrate that our method outperforms state-of-the-art methods in both accuracy and convergence rate.

cs.LG↗
arXiv2023

centroIDA: Cross-Domain Class Discrepancy Minimization Based on Accumulative Class-Centroids for Imbalanced Domain Adaptation

Unsupervised Domain Adaptation (UDA) approaches address the covariate shift problem by minimizing the distribution discrepancy between the source and target domains, assuming that the label distribution is invariant across domains. However, in the imbalanced domain adaptation (IDA) scenario, covariate and long-tailed label shifts both exist across domains. To tackle the IDA problem, some current research focus on minimizing the distribution discrepancies of each corresponding class between source and target domains. Such methods rely much on the reliable pseudo labels' selection and the feature distributions estimation for target domain, and the minority classes with limited numbers makes the estimations more uncertainty, which influences the model's performance. In this paper, we propose a cross-domain class discrepancy minimization method based on accumulative class-centroids for IDA (centroIDA). Firstly, class-based re-sampling strategy is used to obtain an unbiased classifier on source domain. Secondly, the accumulative class-centroids alignment loss is proposed for iterative class-centroids alignment across domains. Finally, class-wise feature alignment loss is used to optimize the feature representation for a robust classification boundary. A series of experiments have proved that our method outperforms other SOTA methods on IDA problem, especially with the increasing degree of label shift.

cs.LG↗
arXiv2023

Dual-Branch Temperature Scaling Calibration for Long-Tailed Recognition

The calibration for deep neural networks is currently receiving widespread attention and research. Miscalibration usually leads to overconfidence of the model. While, under the condition of long-tailed distribution of data, the problem of miscalibration is more prominent due to the different confidence levels of samples in minority and majority categories, and it will result in more serious overconfidence. To address this problem, some current research have designed diverse temperature coefficients for different categories based on temperature scaling (TS) method. However, in the case of rare samples in minority classes, the temperature coefficient is not generalizable, and there is a large difference between the temperature coefficients of the training set and the validation set. To solve this challenge, this paper proposes a dual-branch temperature scaling calibration model (Dual-TS), which considers the diversities in temperature parameters of different categories and the non-generalizability of temperature parameters for rare samples in minority classes simultaneously. Moreover, we noticed that the traditional calibration evaluation metric, Excepted Calibration Error (ECE), gives a higher weight to low-confidence samples in the minority classes, which leads to inaccurate evaluation of model calibration. Therefore, we also propose Equal Sample Bin Excepted Calibration Error (Esbin-ECE) as a new calibration evaluation metric. Through experiments, we demonstrate that our model yields state-of-the-art in both traditional ECE and Esbin-ECE metrics.

cs.LG↗
arXiv2023

Ultra-narrowband interference circuits enable low-noise and high-rate photon counting for InGaAs/InP avalanche photodiodes

Afterpulsing noise in InGaAs/InP single photon avalanche photodiodes (APDs) is caused by carrier trapping and can be suppressed successfully through limiting the avalanche charge via sub-nanosecond gating. Detection of faint avalanches requires an electronic circuit that is able to effectively remove the gate-induced capacitive response while keeping photon signals intact. Here we demonstrate a novel ultra-narrowband interference circuit (UNIC) that can reject the capacitive response by up to 80 dB per stage with little distortion to avalanche signals. Cascading two UNIC's in a readout circuit, we were able to enable high count rate of up to 700 MC/s and low afterpulsing of 0.5 % at a detection efficiency of 25.3 % for 1.25 GHz sinusoidally gated InGaAs/InP APDs. At -30 degree C, we measured 1 % afterpulsing at a detection efficiency of 21.2 %.

quant-ph↗
arXiv2022

Electrically function switchable magnetic domain-wall memory

More-versatile memory is strongly desired for end-users to protect their information in the information era. In particular, bit-level switchable memory, from rewritable to read-only function, allows end-users to prevent any important data from being tampered with. However, no such switchable memory has been reported. We demonstrated the rewritable function can be converted into a read-only function by a sufficiently large current pulse in a U-shaped domain-wall memory composed of an asymmetric Pt/Co/Ru/AlOx heterostructure with strong Dzyaloshinskii-Moriya interaction. Wafer-scale switchable magnetic domain-wall memory arrays on 4-inch Si/SiO2 substrate were designed and fabricated. Furthermore, we confirmed the information can be stored in rewritable or read-only state at bit-level according to the security-needs of end-users. Our work not only provides a solution for personal confidential data, but also paves the way for developing multi-functional spintronic devices.

physics.app-ph↗
arXiv2021

Multi-Factors Aware Dual-Attentional Knowledge Tracing

With the increasing demands of personalized learning, knowledge tracing has become important which traces students' knowledge states based on their historical practices. Factor analysis methods mainly use two kinds of factors which are separately related to students and questions to model students' knowledge states. These methods use the total number of attempts of students to model students' learning progress and hardly highlight the impact of the most recent relevant practices. Besides, current factor analysis methods ignore rich information contained in questions. In this paper, we propose Multi-Factors Aware Dual-Attentional model (MF-DAKT) which enriches question representations and utilizes multiple factors to model students' learning progress based on a dual-attentional mechanism. More specifically, we propose a novel student-related factor which records the most recent attempts on relevant concepts of students to highlight the impact of recent exercises. To enrich questions representations, we use a pre-training method to incorporate two kinds of question information including questions' relation and difficulty level. We also add a regularization term about questions' difficulty level to restrict pre-trained question representations to fine-tuning during the process of predicting students' performance. Moreover, we apply a dual-attentional mechanism to differentiate contributions of factors and factor interactions to final prediction in different practice records. At last, we conduct experiments on several real-world datasets and results show that MF-DAKT can outperform existing knowledge tracing methods. We also conduct several studies to validate the effects of each component of MF-DAKT.

cs.AI↗
arXiv2021

Several special cases of a square problem

Here is a square problem: in a unit square, is there a point with four rational distances to the vertices? A probability argument suggests a negative answer. This paper proves several special cases of the square problem: if the point sits on the diagonals, the midlines or the edges of the square, or the side-length of the square is n times the distance from the point to one side (both $n$ and $(n^2+4)$ are prime numbers), the distances from this point to the four vertices can not be all rational. However, this paper does not prove a more general situation. The proof here can be extended to the whole plane, instead of being limited to the interior of the square. Key Words: discrete Geometry; rational distance; a square problem.

math.GM↗
arXiv2020

Complementary lateral-spin-orbit building blocks for programmable logic and in-memory computing

Current-driven switching of nonvolatile spintronic materials and devices based on spin-orbit torques offer fast data processing speed, low power consumption, and unlimited endurance for future information processing applications. Analogous to conventional CMOS technology, it is important to develop a pair of complementary spin-orbit devices with differentiated magnetization switching senses as elementary building blocks for realizing sophisticated logic functionalities. Various attempts using external magnetic field or complicated stack/circuit designs have been proposed, however, plainer and more feasible approaches are still strongly desired. Here we show that a pair of two locally laser annealed perpendicular Pt/Co/Pt devices with opposite laser track configurations and thereby inverse field-free lateral spin-orbit torques (LSOTs) induced switching senses can be adopted as such complementary spin-orbit building blocks. By electrically programming the initial magnetization states (spin down/up) of each sample, four Boolean logic gates of AND, OR, NAND and NOR, as well as a spin-orbit half adder containing an XOR gate, were obtained. Moreover, various initialization-free, working current intensity-programmable stateful logic operations, including the material implication (IMP) gate, were also demonstrated by regarding the magnetization state as a logic input. Our complementary LSOT building blocks provide an applicable way towards future efficient spin logics and in-memory computing architectures.

cond-mat.mes-hall↗
arXiv2019

Deterministic magnetization switching using lateral spin-orbit torque

Current-induced magnetization switching by spin-orbit torque (SOT) holds considerable promise for next generation ultralow-power memory and logic applications. In most cases, generation of spin-orbit torques has relied on an external injection of out-of-plane spin currents into the magnetic layer, while an external magnetic field along the electric current direction is generally required for realizing deterministic switching by SOT. Here, we report deterministic current-induced SOT full magnetization switching by lateral spin-orbit torque in zero external magnetic field. The Pt/Co/Pt magnetic structure was locally annealed by a laser track along the in-plane current direction, resulting in a lateral Pt gradient within the ferromagnetic layer, as confirmed by microstructure and chemical composition analysis. In zero magnetic field, the direction of the deterministic current-induced magnetization switching depends on the location of the laser track, but shows no dependence on the net polarization of external out-of-plane spin currents. From the behavior under external magnetic fields, we identify two independent mechanisms giving rise to SOT, i.e. the lateral Pt-Co asymmetry as well as out-of-plane injected spin currents, where the polarization and the magnitude of the SOT in the former case depends on the relative location and the laser power of the annealing track. Our results demonstrate an efficient field-free deterministic full magnetization switching scheme, without requiring out-of-plane spin current injection or complex external stack structures.

cond-mat.mes-hall↗
arXiv2019

Spin Logic Devices via Electric Field Controlled Magnetization Reversal by Spin-Orbit Torque

We describe a spin logic device with controllable magnetization switching of perpendicularly magnetized ferromagnet / heavy metal structures on a ferroelectric (1-x)[Pb(Mg1/3Nb2/3)O3]-x[PbTiO3] (PMN-PT) substrate using current-induced spin-orbit torque. The devices were operated without an external magnetic field and controlled by voltages as low as 10 V applied across the PMN-PT substrate, which is much lower compared to previous reports (500 V). The deterministic switching with smaller voltage was realized from the virgin state of the PMN-PT. Ferroelectric simulation shows the unsaturated minor loop exhibits obvious asymmetries in the polarizations. Larger polarization can be induced from the initial ferroelectric state, while it is difficult for opposite polarization. The XNOR, AND, NAND and NOT logic functions were demonstrated by the deterministic magnetization switching from the interaction between the spin-orbit torque and electric field at the PMN-PT/Pt interface. The nonvolatile spin logic scheme in this work is simple, scalable, programmable, which are favorable in the logic-in-memory design with low energy consumption.

physics.app-ph↗
arXiv2018

An Adaptive Oversampling Learning Method for Class-Imbalanced Fault Diagnostics and Prognostics

Data-driven fault diagnostics and prognostics suffers from class-imbalance problem in industrial systems and it raises challenges to common machine learning algorithms as it becomes difficult to learn the features of the minority class samples. Synthetic oversampling methods are commonly used to tackle these problems by generating the minority class samples to balance the distributions between majority and minority classes. However, many of oversampling methods are inappropriate that they cannot generate effective and useful minority class samples according to different distributions of data, which further complicate the process of learning samples. Thus, this paper proposes a novel adaptive oversampling technique: EM-based Weighted Minority Oversampling TEchnique (EWMOTE) for industrial fault diagnostics and prognostics. The methods comprises a weighted minority sampling strategy to identify hard-to-learn informative minority fault samples and Expectation Maximization (EM) based imputation algorithm to generate fault samples. To validate the performance of the proposed methods, experiments are conducted in two real datasets. The results show that the method could achieve better performance on not only binary class, but multi-class imbalance learning task in different imbalance ratios than other oversampling-based baseline models.

cs.LG↗
arXiv2018

Anomalous Hall effect induced spin Hall magnetoresistance in an antiferromagnetic Cr2O3/Ta bilayer

The spin Hall magnetoresistance (SMR) and anomalous Hall effect (AHE) are observed in a Cr2O3/Ta structure. The structural and surface morphology of Cr2O3/Ta bilayers have been investigated. Temperature dependence of longitudinal and transverse resistances measurements confirm the relationship between SMR and AHE signals in Cr2O3/Ta structure. By means of temperature dependent magnetoresistance measurements, the physical origin of SMR in the Cr2O3/Ta structure is revealed, and the contribution to the SMR from the spin current generated by AHE has been proved. The so-called boundary magnetization due to the bulk antiferromagnetic order in Cr2O3 film may be responsible for the relationship of SMR and AHE in the Cr2O3/Ta bilayer.

cond-mat.mtrl-sci↗
arXiv2016

Piezo Voltage Controlled Planar Hall Effect Devices

The electrical control of the magnetization switching in ferromagnets is highly desired for future spintronic applications. Here we report on hybrid piezoelectric (PZT) /ferromagnetic (Co2FeAl) devices in which the planar Hall voltage in the ferromagnetic layer is tuned solely by piezo voltages. The change of planar Hall voltage is associated with magnetization switching through 90 in the plane under piezo voltages. Room temperature magnetic NOT and NOR gates are demonstrated based on the piezo voltage controlled Co2FeAl planar Hall effect devices without the external magnetic field. Our demonstration may lead to the realization of both information storage and processing using ferromagnetic materials.

cond-mat.mes-hall↗
arXiv2016

Electric field control of deterministic current-induced magnetization switching in a hybrid ferromagnetic/ferroelectric structure

All-electrical and programmable manipulations of ferromagnetic bits are highly pursued for the aim of high integration and low energy consumption in modern information technology. Methods based on the spin-orbit torque switching in heavy metal/ferromagnet structures have been proposed with magnetic field, and recently are heading toward deterministic switching without external magnetic field. Here we demonstrate that an in-plane effective magnetic field can be induced by an electric field without breaking the symmetry of the structure of the thin film, and realize the deterministic magnetization switching in a hybrid ferromagnetic/ferroelectric structure with Pt/Co/Ni/Co/Pt layers on PMN-PT substrate. The effective magnetic field can be reversed by changing the direction of the applied electric field on the PMN-PT substrate, which fully replaces the controllability function of the external magnetic field. The electric field is found to generate an additional spin-orbit torque on the CoNiCo magnets, which is confirmed by macrospin calculations.

cond-mat.mtrl-sci↗