arXiv subjects

Yawen Zhao

Publications and source records attributed to Yawen Zhao.

7 recordsLinked to original sources

arXiv2025

Machine Unlearning for Streaming Forgetting

Machine unlearning aims to remove knowledge of the specific training data in a well-trained model. Currently, machine unlearning methods typically handle all forgetting data in a single batch, removing the corresponding knowledge all at once upon request. However, in practical scenarios, requests for data removal often arise in a streaming manner rather than in a single batch, leading to reduced efficiency and effectiveness in existing methods. Such challenges of streaming forgetting have not been the focus of much research. In this paper, to address the challenges of performance maintenance, efficiency, and data access brought about by streaming unlearning requests, we introduce a streaming unlearning paradigm, formalizing the unlearning as a distribution shift problem. We then estimate the altered distribution and propose a novel streaming unlearning algorithm to achieve efficient streaming forgetting without requiring access to the original training data. Theoretical analyses confirm an $O(\sqrt{T} + V_T)$ error bound on the streaming unlearning regret, where $V_T$ represents the cumulative total variation in the optimal solution over $T$ learning rounds. This theoretical guarantee is achieved under mild conditions without the strong restriction of convex loss function. Experiments across various models and datasets validate the performance of our proposed method.

cs.LG↗

arXiv2024

GENIU: A Restricted Data Access Unlearning for Imbalanced Data

With the increasing emphasis on data privacy, the significance of machine unlearning has grown substantially. Class unlearning, which involves enabling a trained model to forget data belonging to a specific class learned before, is important as classification tasks account for the majority of today's machine learning as a service (MLaaS). Retraining the model on the original data, excluding the data to be forgotten (a.k.a forgetting data), is a common approach to class unlearning. However, the availability of original data during the unlearning phase is not always guaranteed, leading to the exploration of class unlearning with restricted data access. While current unlearning methods with restricted data access usually generate proxy sample via the trained neural network classifier, they typically focus on training and forgetting balanced data. However, the imbalanced original data can cause trouble for these proxies and unlearning, particularly when the forgetting data consists predominantly of the majority class. To address this issue, we propose the GENerative Imbalanced Unlearning (GENIU) framework. GENIU utilizes a Variational Autoencoder (VAE) to concurrently train a proxy generator alongside the original model. These generated proxies accurately represent each class and are leveraged in the unlearning phase, eliminating the reliance on the original training data. To further mitigate the performance degradation resulting from forgetting the majority class, we introduce an in-batch tuning strategy that works with the generated proxies. GENIU is the first practical framework for class unlearning in imbalanced data settings and restricted data access, ensuring the preservation of essential information for future unlearning. Experimental results confirm the superiority of GENIU over existing methods, establishing its effectiveness in empirical scenarios.

cs.LG↗

arXiv2024

Label-Agnostic Forgetting: A Supervision-Free Unlearning in Deep Models

Machine unlearning aims to remove information derived from forgotten data while preserving that of the remaining dataset in a well-trained model. With the increasing emphasis on data privacy, several approaches to machine unlearning have emerged. However, these methods typically rely on complete supervision throughout the unlearning process. Unfortunately, obtaining such supervision, whether for the forgetting or remaining data, can be impractical due to the substantial cost associated with annotating real-world datasets. This challenge prompts us to propose a supervision-free unlearning approach that operates without the need for labels during the unlearning process. Specifically, we introduce a variational approach to approximate the distribution of representations for the remaining data. Leveraging this approximation, we adapt the original model to eliminate information from the forgotten data at the representation level. To further address the issue of lacking supervision information, which hinders alignment with ground truth, we introduce a contrastive loss to facilitate the matching of representations between the remaining data and those of the original model, thus preserving predictive performance. Experimental results across various unlearning tasks demonstrate the effectiveness of our proposed method, Label-Agnostic Forgetting (LAF) without using any labels, which achieves comparable performance to state-of-the-art methods that rely on full supervision information. Furthermore, our approach excels in semi-supervised scenarios, leveraging limited supervision information to outperform fully supervised baselines. This work not only showcases the viability of supervision-free unlearning in deep models but also opens up a new possibility for future research in unlearning at the representation level.

cs.LG↗

arXiv2022

A Boosting Algorithm for Positive-Unlabeled Learning

Positive-unlabeled (PU) learning deals with binary classification problems when only positive (P) and unlabeled (U) data are available. Many recent PU methods are based on neural networks, but little has been done to develop boosting algorithms for PU learning, despite boosting algorithms' strong performance on many fully supervised classification problems. In this paper, we propose a novel boosting algorithm, AdaPU, for PU learning. Similarly to AdaBoost, AdaPU aims to optimize an empirical exponential loss, but the loss is based on the PU data, rather than on positive-negative (PN) data. As in AdaBoost, we learn a weighted combination of weak classifiers by learning one weak classifier and its weight at a time. However, AdaPU requires a very different algorithm for learning the weak classifiers and determining their weights. This is because AdaPU learns a weak classifier and its weight using a weighted positive-negative (PN) dataset with some negative data weights $-$ the dataset is derived from the original PU data, and the data weights are determined by the current weighted classifier combination, but some data weights are negative. Our experiments showed that AdaPU outperforms neural networks on several benchmark PU datasets, including a large-scale challenging cyber security dataset.

cs.LG↗

arXiv2021

First-principles study of the crystal structure, electronic structure, and transport properties of NiTe$_2$ under pressure

Recent experiments showed the distinct observations on the transition metal ditelluride NiTe$_2$ under pressure: one reported a superconducting phase transition at 12 GPa, whereas another observed a sign reversal of Hall resistivity at 16 GPa without the appearance of superconductivity. To clarify the controversial experimental phenomena, we have carried out first-principles electronic structure calculations on the compressed NiTe$_2$ with structure searching and optimization. Our calculations show that the pressure can transform NiTe$_2$ from a layered P-3m1 phase to a cubic Pa-3 phase at $\sim$10 GPa. Meanwhile, both the P-3m1 and Pa-3 phases possess nontrivial topological properties. The calculated superconducting $T_c$'s for these two phases based on the electron-phonon coupling theory both approach 0 K. Further magnetic transport calculations reveal that the sign of Hall resistance for the Pa-3 phase is sensitive to the pressure and the charge doping, in contrast to the case of the P-3m1 phase. Our theoretical predictions on the compressed NiTe$_2$ wait for careful experimental examinations.

cond-mat.supr-con↗

arXiv2020

Interface reflectivity of a superdiffusive spin current in ultrafast demagnetization and THz emission

The spin- and energy-dependent interface reflectivity of a ferromagnetic (FM) film in contact with a nonmagnetic (NM) film is calculated using a first-principles transport method and incorporated into the superdiffusive spin transport model to study the femtosecond laser-induced ultrafast demagnetization of Fe|NM and Ni|NM (NM= Au, Al & Pt) bilayers. By comparing the calculated demagnetization with transparent and real interfaces, we demonstrate that the spin-dependent reflection of hot electrons has a noticeable influence on the ultrafast demagnetization and the associated terahertz electromagnetic radiation. In particular, a spin filtering effect is found at the Fe|NM interface that increases the spin current injected into the NM metal, which enhances both the resulting demagnetization and the resulting THz emission. This suggests that the THz radiation can be optimized by tailoring the interface, indicating a very large tunability.

cond-mat.mes-hall↗

arXiv2018

Highly efficient visible colloidal lead-halide perovskite nanocrystal light-emitting diodes

Lead-halide perovskites have been attracting attention for potential use in solid-state lighting. Following the footsteps of solar cells, the field of perovskite light-emitting diodes (PeLEDs) has been growing rapidly. Their application prospects in lighting, however, remain still uncertain due to a variety of shortcomings in device performance including their limited levels of luminous efficiency achievable thus far. Here we show high-efficiency PeLEDs based on colloidal perovskite nanocrystals (PeNCs) synthesized at room temperature possessing dominant first-order excitonic radiation (enabling a photoluminescence quantum yield of 71% in solid film), unlike in the case of bulk perovskites with slow electron-hole bimolecular radiative recombination (a second-order process). In these PeLEDs, by reaching charge balance in the recombination zone, we find that the Auger nonradiative recombination, with its significant role in emission quenching, is effectively suppressed in low driving current density range. In consequence, these devices reach a record high maximum external quantum efficiency of 12.9% reported to date and an unprecedentedly high power efficiency of 30.3 lm W-1 at luminance levels above 1000 cd m-2 as required for various applications. These findings suggest that, with feasible levels of device performance, the PeNCs hold great promise for their use in LED lighting and displays.

physics.app-ph↗