49 citations · 105 across the 22 of their papers we have counts for
23 papers
Trust your neighbours: Penalty-based constraints for model calibration
Balamurali Murugesan, Sukesh Adiga, Bingyuan Liu +3
Ensuring reliable confidence scores from deep networks is of pivotal importance in critical decision-making systems, notably in the medical domain. While recent literature on calib…
Class Adaptive Network Calibration
Bingyuan Liu, Jérôme Rony, Adrian Galdran +2
Recent studies have revealed that, beyond conventional accuracy, calibration should also be considered for training modern deep neural networks. To address miscalibration during le…
Diederich--Fornæss index and global regularity in the --Neumann problem: domains with comparable Levi eigenvalues
Bingyuan Liu, Emil J. Straube
Let be a smooth bounded pseudoconvex domain in . Let . We show that if --sums of eigenvalues of the Levi form are comparable, then…
GAMMA Challenge:Glaucoma grAding from Multi-Modality imAges
Junde Wu, Huihui Fang, Fei Li +26
Color fundus photography and Optical Coherence Tomography (OCT) are the two most cost-effective tools for glaucoma screening. Both two modalities of images have prominent biomarker…
The Devil is in the Margin: Margin-based Label Smoothing for Network Calibration
Bingyuan Liu, Ismail Ben Ayed, Adrian Galdran +1
In spite of the dominant performances of deep neural networks, recent works have shown that they are poorly calibrated, resulting in over-confident predictions. Miscalibration can…
Robust High-Dimensional Regression with Coefficient Thresholding and its Application to Imaging Data Analysis
Bingyuan Liu, Qi Zhang, Lingzhou Xue +2
It is of importance to develop statistical techniques to analyze high-dimensional data in the presence of both complex dependence and possible outliers in real-world applications s…
Segmentation with mixed supervision: Confidence maximization helps knowledge distillation
Bingyuan Liu, Christian Desrosiers, Ismail Ben Ayed +1
Despite achieving promising results in a breadth of medical image segmentation tasks, deep neural networks require large training datasets with pixel-wise annotations. Obtaining th…
The -Neumann problem and boundary integral equations
Bingyuan Liu
In this note, we find an equivalent boundary integral equation to the classical -Neumann problem. The new equation contains an equivalent regularity to the global r…
Do We Really Need Dice? The Hidden Region-Size Biases of Segmentation Losses
Bingyuan Liu, Jose Dolz, Adrian Galdran +2
Most segmentation losses are arguably variants of the Cross-Entropy (CE) or Dice losses. On the surface, these two categories of losses seem unrelated, and there is no clear consen…
Improving Neural Network Robustness through Neighborhood Preserving Layers
Bingyuan Liu, Christopher Malon, Lingzhou Xue +1
Robustness against adversarial attack in neural networks is an important research topic in the machine learning community. We observe one major source of vulnerability of neural ne…
A Manifold Proximal Linear Method for Sparse Spectral Clustering with Application to Single-Cell RNA Sequencing Data Analysis
Zhongruo Wang, Bingyuan Liu, Shixiang Chen +3
Spectral clustering is one of the fundamental unsupervised learning methods widely used in data analysis. Sparse spectral clustering (SSC) imposes sparsity to the spectral clusteri…
Furnishing Your Room by What You See: An End-to-End Furniture Set Retrieval Framework with Rich Annotated Benchmark Dataset
Bingyuan Liu, Jiantao Zhang, Xiaoting Zhang +3
Understanding interior scenes has attracted enormous interest in computer vision community. However, few works focus on the understanding of furniture within the scenes and a large…
The Diederich--Fornæss index and the regularities on the -Neumann problem
Bingyuan Liu
We show, under an assumption on the weakly pseudoconvex points, the trivial Diederich--Fornæss index directly implies the global regularities of the -Neumann operat…
The -Neumann operator with the Sobolev norm of integer orders
Phillip Harrington, Bingyuan Liu
Let be a bounded pseudoconvex domain with smooth boundary. For each , we give a sufficient condition to estimate the -Neumann o…
The Diederich--Fornæss index II: for domains of trivial index
Bingyuan Liu
We study bounded pseudoconvex domains in complex Euclidean spaces. We find analytical necessary conditions and geometric sufficient conditions for a domain being of trivial Diederi…
The Diederich-Fornæss index I: for domains of non-trivial index
Bingyuan Liu
We study bounded pseudoconvex domains in complex Euclidean space. We define an index associated to the boundary and show this new index is equivalent to the Diederich-Fornæss index…
The intrinsic geometry on bounded pseudoconvex domains
Bingyuan Liu
The Diederich--Fornæss index has been introduced since 1977 to classify bounded pseudoconvex domains. In this article, we derive several intrinsic, geometric conditions on boundary…
Geometric Analysis on the Diederich-Fornæss Index
Steven G. Krantz, Bingyuan Liu, Marco Peloso
We derive a sufficient condition on a bounded pseudoconvex domain with smooth boundary such that is plurisubharmonic on for arbitrarily c…
The limit set for discrete complex hyperbolic groups
Angel Cano, Bingyuan Liu, Marlon M. López
Given a discrete subgroup of it acts by isometries on the unit complex ball , in this setting a lot of work has been done in order to understand…
Two applications of the Schwarz lemma
Bingyuan Liu
The Schwarz lemmas are well-known characterizations for holomorphic maps and we exhibit two examples of their applications. For a sequence family of biholomorphisms , it is us…
Analysis of orbit accumulation points and the Greene-Krantz conjecture
Bingyuan Liu
In , we classify the domains for which is noncompact and describe these domains by their defining functions. This note is based on the technique of the s…
The Wong-Rosay type theorem for Kähler manifolds
Bingyuan Liu
The Wong-Rosay theorem characterizes the strongly pseudoconvex domains of by their automorphism groups. It has a lot of generalizations to other kinds of domains (fo…
Finite type domains with hyperbolic orbit accumulation points
Bingyuan Liu
In this paper, finite type domains with hyperbolic orbit accumulation points are studied. We prove, in case of , it has to be a (global) pseudoconvex domain, after an…