Γ and B

By L. Sehgal, J. V. Leusen. 2004

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Natural Language Processing

Citations in the corpus

Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks (2015, background): “It is noteworthy that the Recall-to-IoU metric is just loosely [9, 8, 1] related to the ultimate detection accuracy.”
Evaluating the Social Impact of Generative AI Systems in Systems and Society (2023, background): “Tools [215] are being developed to protect certain modalities from being used as training data.”
Panoptic Segmentation (2018, background): “Stuff classifiers are usually built on fully convolutional nets [30] with dilations [51, 5] while object detectors often use object proposals [15] and are region-based [37, 14].”
Learning to Segment Object Candidates (2015, methodology): “These methods achieve top results on object detection (when coupled with R-CNNs [10]) and also obtain the best AR [13].”
BIVA: A Very Deep Hierarchy of Latent Variables for Generative Modeling (2019, methodology): “Based on a thorough study, [34] states that the main issue is the fact that explicit density models tend to capture low-level statistics, as opposed to the high-level semantics that are preferable when doing anomaly detection.”
A Note on Data Biases in Generative Models (2020, methodology): “We interpret this task as a minimization of their mutual information I(z, y), and, following [6], minimize an upper bound on it where we model the distribution p(z|y) with a conditional INN τ and use a standard normal prior q(z).”
A MultiPath Network for Object Detection (2016, methodology): “Object Proposals: When originally introduced, object proposals were based on lowlevel grouping cues, edges, and superpixels [1, 2, 16, 31, 33].”
How To Backdoor Federated Learning (2018, methodology): “Finally, their attack works only against a single-layer feed-forward network or CNN and does not converge for large networks such as the original federated learning framework [43].”
Byzantine-Resilient Secure Federated Learning (2020, methodology): “CIFAR-10 dataset: To investigate the performance of BREA with a larger dataset, we additionally experiment with CIFAR10 [43], by using the same model architecture of [3] (about 106 model parameters).”
Differential Privacy Has Disparate Impact on Model Accuracy (2019, methodology): “User-level DP in federated learning [24] was demonstrated in [16, 37].”
Contamination Attacks and Mitigation in Multi-Party Machine Learning (2018, background): “Additionally, it allows parties to outsource potentially heavy computation and guarantees that they do not see model parameters during training that have to be shared, for example, in distributed settings [16, 27, 35, 36, 39].”
DeepFool: A Simple and Accurate Method to Fool Deep Neural Networks (2015, neutral): “In [14], based on the experiments done on Pascal3D+ annotations, the authors showed that convolutional networks are not invariant to some sort of transformations.”
A Survey of Methods for Explaining Black Box Models (2018, methodology): “Therefore, in data mining and machine learning, interpretability is defined as the ability to explain or to provide the meaning in understandable terms to a human [24].”
A Kernel Test for Three-Variable Interactions (2013, background): “, for all topological spaces where it is possible to construct a valid positive definite function; see [23] for details.”
Multi-view 3D Object Detection Network for Autonomous Driving (2016, methodology): “Different from 2D box recall [13], we compute the IoU overlap of two cuboids.”
Causal Inference Using the Algorithmic Markov Condition (2008, background): “[15]) such that the author of the text that appeared later is blamed to have copied it from t he other one or both are blamed to have copied from a third one.”
Testing for Homogeneity with Kernel Fisher Discriminant Analysis (2007, background): “The following proposition shows that the limit is finite, str ic ly positive and independent of the kernel otherwise (see [8] for similar results for canonical correl ation analysis).”
A Dataset and Exploration of Models for Understanding Video Data through Fill-in-the-Blank Question-Answering (2016, methodology): “This dataset was used as the basis of the Large Scale Movie Description Challenge (LSMDC) in 2015 and 2016 [30].”