Author name cluster

Yuqi Feng

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

4 papers

2 author rows

YNIMG Journal 2025 Journal Article

Mesoscale functional connectivity of amygdala to the auditory and prefrontal cortex of macaque monkeys revealed by INS-fMRI

Qianbing Li
An Ping
Yuqi Feng
Bin Xu
Baorong Zhang
Anna Wang Roe
Lixia Gao
Xinjian Li

Mammals rely heavily on their auditory system to perceive environmental threats, socially communicate, and care for the young. As an extension of the multiple sensory system including the auditory system, the amygdala evaluates the emotional salience of acoustic stimuli, and mediates its impact on sensory, cognitive, and physiological aspects of emotional processing via the lateral amygdala (LA), basal amygdala (BA), and central amygdala (CeA) nuclei of the amygdala in acoustic domain. However, the functional connections of LA, BA, and CeA with the auditory cortex (AC) and the prefrontal cortex (PFC) remain unclear, particularly at the mesoscale level. Here we employed a novel method called INS-fMRI (Infrared Neural Stimulation combined with high-resolution functional magnetic resonance imaging) in Macaque monkeys, this method permits stimulation of multiple sites within single animals in vivo, so that the relative organization of auditory networks can be studied. We found that: (1) Focal INS stimulation of the amygdala elicited robust and reliable responses in both the AC and the PFC; (2) Amygdala stimulation mainly activated ipsilateral AC and PFC; (3) The stimulation of the amygdala mainly activated the secondary AC, and the dorsolateral PFC; (4) The connection between the amygdala and the cortex is mainly mediated by neurons in LA and BA connection area. Our study further revealed the functional connectivity among the amygdala subnucleus, the auditory cortex and the prefrontal cortex, and will shed light on the research for processing biologically meaningful complex sounds.

Details DOI

NeurIPS Conference 2025 Conference Paper

Vulnerable Data-Aware Adversarial Training

Yuqi Feng
Jiahao Fan
Yanan Sun

Fast adversarial training (FAT) has been considered as one of the most effective alternatives to the computationally-intensive adversarial training. Generally, FAT methods pay equal attention to each sample of the target task. However, the distance between each sample and the decision boundary is different, learning samples which are far from the decision boundary (i. e. , less important to adversarial robustness) brings additional training cost and leads to sub-optimal results. To tackle this issue, we present vulnerable data-aware adversarial training (VDAT) in this study. Specifically, we first propose a margin-based vulnerability calculation method to measure the vulnerability of data samples. Moreover, we propose a vulnerability-aware data filtering method to reduce the training data for adversarial training thus improve the training efficiency. The experiments are conducted in terms of adversarial training and robust neural architecture search on CIFAR-10, CIFAR-100, and ImageNet-1K. The results demonstrate that VDAT is up to 76% more efficient than state-of-the-art FAT methods, while achieving improvements regarding the natural accuracy and adversarial accuracy in both scenarios. Furthermore, the visualizations and ablation studies show the effectiveness of both core components designed in VDAT.

PDF Details

ICLR Conference 2025 Conference Paper

Zero-cost Proxy for Adversarial Robustness Evaluation

Yuqi Feng
Yuwei Ou
Jiahao Fan
Yanan Sun 0001

Deep neural networks (DNNs) easily cause security issues due to the lack of adversarial robustness. An emerging research topic for this problem is to design adversarially robust architectures via neural architecture search (NAS), i.e., robust NAS. However, robust NAS needs to train numerous DNNs for robustness estimation, making the search process prohibitively expensive. In this paper, we propose a zero-cost proxy to evaluate the adversarial robustness without training. Specifically, the proposed zero-cost proxy formulates the upper bound of adversarial loss, which can directly reflect the adversarial robustness. The formulation involves only the initialized weights of DNNs, thus the training process is no longer needed. Moreover, we theoretically justify the validity of the proposed proxy based on the theory of neural tangent kernel and input loss landscape. Experimental results show that the proposed zero-cost proxy can bring more than $20\times$ speedup compared with the state-of-the-art robust NAS methods, while the searched architecture has superior robustness and transferability under white-box and black-box attacks. Furthermore, compared with the state-of-the-art zero-cost proxies, the calculation of the proposed method has the strongest correlation with adversarial robustness. Our source code is available at https://github.com/fyqsama/Robust_ZCP.

Details

IJCAI Conference 2024 Conference Paper

CAP: A Context-Aware Neural Predictor for NAS

Han Ji
Yuqi Feng
Yanan Sun

Neural predictors are effective in boosting the time-consuming performance evaluation stage in neural architecture search (NAS), owing to their direct estimation of unseen architectures. Despite the effectiveness, training a powerful neural predictor with fewer annotated architectures remains a huge challenge. In this paper, we propose a context-aware neural predictor (CAP) which only needs a few annotated architectures for training based on the contextual information from the architectures. Specifically, the input architectures are encoded into graphs and the predictor infers the contextual structure around the nodes inside each graph. Then, enhanced by the proposed context-aware self-supervised task, the pre-trained predictor can obtain expressive and generalizable representations of architectures. Therefore, only a few annotated architectures are sufficient for training. Experimental results in different search spaces demonstrate the superior performance of CAP compared with state-of-the-art neural predictors. In particular, CAP can rank architectures precisely at the budget of only 172 annotated architectures in NAS-Bench-101. Moreover, CAP can help find promising architectures in both NAS-Bench-101 and DARTS search spaces on the CIFAR-10 dataset, serving as a useful navigator for NAS to explore the search space efficiently.

PDF Details DOI