John E. Hopcroft

John E. Hopcroft

Auto-generated profile, based on publicly available information

John E. Hopcroft

Cornell University, Ithaca, USA

Research Focus & Metrics

Academic Career Length

15Years (2012-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 April 2017 in arXiv
Verified
Preprint
Open access
PDF file
GHGao Huang
Gao Huang
YLYixuan Li
Yixuan Li
GPGeoff Pleiss
Geoff Pleiss
ZLZhuang Liu
Zhuang Liu
JHJohn E. Hopcroft
John E. Hopcroft
KWKilian Q. Weinberger
Kilian Q. Weinberger
Ensembles of neural networks are known to be much more robust and accurate than individual networks. However, training multiple deep networks for model averaging is computationally expensive. In this paper, we propose a method to obtain the seemingly contradictory goal of ensembling multiple neural networks at no additional training cost. We achieve this goal by training a single neural network, converging to several local minima along its optimization path and saving the model parameters. To obtain repeated rapid convergence, we leverage recent work on cyclic learning rate schedules. The resulting technique, which we refer to as Snapshot Ensembling, is simple, yet surprisingly effective. We show in a series of experiments that our approach is compatible with diverse network architectures and learning tasks. It consistently yields lower error rates than state-of-the-art single models at no additional training cost, and compares favorably with traditional network ensembles. On CIFAR-10 and CIFAR-100 our DenseNet Snapshot Ensembles obtain error rates of 3.4% and 17.4% respectively.

24 January 2025 in IEEE Transactions on Information Forensics and Security
Verified
Journal Article
Closed access
XLXin Liu
Xin Liu
YYYichen Yang
Yichen Yang
KHKun He
Kun He
JHJohn E. Hopcroft
John E. Hopcroft
Though deep neural networks exhibit superior performance on various tasks, they are still plagued by adversarial examples. Adversarial training has been demonstrated to be the most effective method to defend against adversarial attacks. However, existing adversarial training methods show that the model robustness has apparent oscillations and overfitting issues in the training process, degrading the defense efficacy. To address these issues, we propose a novel framework called Parameter Interpolation Adversarial Training (PIAT). PIAT tunes the model parameters between each epoch by interpolating the parameters of the previous and current epochs. It makes the decision boundary of model change more moderate and alleviates the overfitting issue, helping the model converge better and achieving higher model robustness. In addition, we suggest using the Normalized Mean Square Error (NMSE) to further improve the robustness by aligning the relative magnitude of logits between clean and adversarial examples rather than the absolute magnitude. Extensive experiments conducted on several benchmark datasets demonstrate that our framework could prominently improve the robustness of both Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs).

31 December 2012 in Internet Mathematics
Verified
Journal Article
Closed access
PDF file
LWLiaoruo Wang
Liaoruo Wang
JHJohn Hopcroft
John Hopcroft
JHJing He
Jing He
HLHongyu Liang
Hongyu Liang
SSSupasorn Suwajanakorn
Supasorn Suwajanakorn
An (α, β)-community is a connected subgraph _C_ with each vertex in _C_ connected to at least β vertices of _C_ (self-loops counted) and each vertex outside of _C_ connected to at most α vertices of _C_ (α<β). In this paper, we present a heuristic algorithm that in practice successfully finds a fundamental community structure. We also explore the structure of (α, β)-communities in various social networks. (α, β)-communities are well clustered into a small number of disjoint groups, and there are no isolated (α, β)-communities scattered between these groups. Two (α, β)-communities in the same group have significant overlap, while those in different groups have extremely small resemblance. A surprising core structure is discovered by taking the intersection of each group of massively overlapping (α, β)-communities. Further, similar experiments on random graphs demonstrate that the core structure found in many social networks is due to their underlying social structure, rather than to high-degree vertices or a particular degree distribution.

3 February 2020 in arXiv
Verified
Preprint
Open access
PDF file
JLJiadong Lin
Jiadong Lin
CSChuanbiao Song
Chuanbiao Song
KHKun He
Kun He
LWLiwei Wang
Liwei Wang
JHJohn E. Hopcroft
John E. Hopcroft
Deep learning models are vulnerable to adversarial examples crafted by applying human-imperceptible perturbations on benign inputs. However, under the black-box setting, most existing adversaries often have a poor transferability to attack other defense models. In this work, from the perspective of regarding the adversarial example generation as an optimization process, we propose two new methods to improve the transferability of adversarial examples, namely Nesterov Iterative Fast Gradient Sign Method (NI-FGSM) and Scale-Invariant attack Method (SIM). NI-FGSM aims to adapt Nesterov accelerated gradient into the iterative attacks so as to effectively look ahead and improve the transferability of adversarial examples. While SIM is based on our discovery on the scale-invariant property of deep learning models, for which we leverage to optimize the adversarial perturbations over the scale copies of the input images so as to avoid "overfitting" on the white-box model being attacked and generate more transferable adversarial examples. NI-FGSM and SIM can be naturally integrated to build a robust gradient-based attack to generate more transferable adversarial examples against the defense models. Empirical results on ImageNet dataset demonstrate that our attack methods exhibit higher transferability and achieve higher attack success rates than state-of-the-art gradient-based attacks.

27 February 2023 in ACM Transactions on Knowledge Discovery From Data
Verified
Journal Article
Open access
MWMeng Wang
Meng Wang
BLBoyu Li
Boyu Li
KHKun He
Kun He
JHJohn E. Hopcroft
John E. Hopcroft
Hidden community is a useful concept proposed recently for social network analysis. Hidden communities indicate some weak communities whose most members also belong to other stronger dominant communities. Dominant communities could form a layer that partitions all the individuals of a network, and hidden communities could form other layer(s) underneath. These layers could be natural structures in the real-world networks like students grouped by major, minor, hometown, and so on. To handle the rapid growth of network scale, in this work, we explore the detection of hidden communities from the local perspective, and propose a new method that detects and boosts each layer iteratively on a subgraph sampled from the original network. We first expand the seed set from a single seed node based on our modified local spectral method and detect an initial dominant local community. Then we temporarily remove the members of this community as well as their connections to other nodes, and detect all the neighborhood communities in the remaining subgraph, including some “broken communities” that only contain a fraction of members in the original network. The local community and neighborhood communities form a dominant layer, and by reducing the edge weights inside these communities, we weaken this layer’s structure to reveal the hidden layers. Eventually, we repeat the whole process, and all communities containing the seed node can be detected and boosted iteratively. We theoretically show that our method can avoid some situations that a broken community and the local community are regarded as one community in the subgraph, leading to the inaccuracy of detection which can be caused by global hidden community detection methods. Extensive experiments show that our method could significantly outperform the state-of-the-art baselines designed for either global hidden community detection or multiple local community detection.

23 January 2015 in arXiv
Verified
Preprint
Open access
PDF file
KHKun He
Kun He
SSSucheta Soundarajan
Sucheta Soundarajan
XCXuezhi Cao
Xuezhi Cao
JHJohn Hopcroft
John Hopcroft
MHMenglong Huang
Menglong Huang
We introduce a new conception of community structure, which we refer to as hidden community structure. Hidden community structure refers to a specific type of overlapping community structure, in which the detection of weak, but meaningful, communities is hindered by the presence of stronger communities. We present Hidden Community Detection HICODE, an algorithm template that identifies both the strong, dominant community structure as well as the weaker, hidden community structure in networks. HICODE begins by first applying an existing community detection algorithm to a network, and then removing the structure of the detected communities from the network. In this way, the structure of the weaker communities becomes visible. Through application of HICODE, we demonstrate that a wide variety of real networks from different domains contain many communities that, though meaningful, are not detected by any of the popular community detection algorithms that we consider. Additionally, on both real and synthetic networks containing a hidden ground-truth community structure, HICODE uncovers this structure better than any baseline algorithms that we compared against. For example, on a real network of undergraduate students that can be partitioned either by `Dorm' (residence hall) or `Year', we see that HICODE uncovers the weaker `Year' communities with a JCRecall score (a recall-based metric that we define in the text) of over 0.7, while the baseline algorithms achieve scores below 0.2.