All Research Areas
Research Areas
Year Published

220 Results

July 7, 2018

Algorithmic Regularization in Over-parameterized Matrix Sensing and Neural Networks with Quadratic Activations


We show that the gradient descent algorithm provides an implicit regularization effect in the learning of over-parameterized matrix factorization models and one-hidden-layer neural networks with quadratic activations.

By: Yuanzhi Li, Tengyu Ma, Hongyang Zhang
June 29, 2018

Understanding the Loss Surface of Neural Networks for Binary Classification

International Conference on Machine Learning (ICML)

Here we focus on the training performance of neural networks for binary classification, and provide conditions under which the training error is zero at all local minima of appropriately chosen surrogate loss functions.

By: Shiyu Liang, Ruoyu Sun, Yixuan Li, R. Srikant
June 19, 2018

LAMV: Learning to align and match videos with kernelized temporal layers

Computer Vision and Pattern Recognition (CVPR)

This paper considers a learnable approach for comparing and aligning videos. Our architecture builds upon and revisits temporal match kernels within neural networks: we propose a new temporal layer that finds temporal alignments by maximizing the scores between two sequences of vectors, according to a time-sensitive similarity metric parametrized in the Fourier domain.

By: Lorenzo Baraldi, Matthijs Douze, Rita Cucchiara, Hervé Jégou
June 19, 2018

Link and code: Fast indexing with graphs and compact regression codes

Computer Vision and Pattern Recognition (CVPR)

Similarity search approaches based on graph walks have recently attained outstanding speed-accuracy trade-offs, taking aside the memory requirements. In this paper, we revisit these approaches by considering, additionally, the memory constraint required to index billions of images on a single server.

By: Matthijs Douze, Alexandre Sablayrolles, Hervé Jégou
June 19, 2018

A Generative Adversarial Approach for Zero-Shot Learning from Noisy Texts

Computer Vision and Pattern Recognition (CVPR)

Most existing zero-shot learning methods consider the problem as a visual semantic embedding one. Given the demonstrated capability of Generative Adversarial Networks (GANs) to generate images, we instead leverage GANs to imagine unseen categories from text descriptions and hence recognize novel classes with no examples being seen.

By: Yizhe Zhu, Mohamed Elhoseiny, Bingchen Liu, Xi Peng, Ahmed Elgammal
June 18, 2018

Improving Landmark Localization with Semi-Supervised Learning

Computer Vision and Pattern Recognition (CVPR)

We present two techniques to improve landmark localization in images from partially annotated datasets. Our primary goal is to leverage the common situation where precise landmark locations are only provided for a small data subset, but where class labels for classification or regression tasks related to the landmarks are more abundantly available.

By: Sina Honari, Pavlo Molchanov, Stephen Tyree, Pascal Vincent, Christopher Pal, Jan Kautz
June 18, 2018

Multimodal Explanations: Justifying Decisions and Pointing to the Evidence

Computer Vision and Pattern Recognition (CVPR)

Deep models that are both effective and explainable are desirable in many settings; prior explainable models have been unimodal, offering either image-based visualization of attention weights or text-based generation of post-hoc justifications. We propose a multimodal approach to explanation, and argue that the two modalities provide complementary explanatory strengths.

By: Dong Huk Park, Lisa Anne Hendricks, Zeynep Akata, Anna Rohrbach, Bernt Schiele, Trevor Darrell, Marcus Rohrbach
June 18, 2018

Don’t Just Assume; Look and Answer: Overcoming Priors for Visual Question Answering

Computer Vision and Pattern Recognition (CVPR)

A number of studies have found that today’s Visual Question Answering (VQA) models are heavily driven by superficial correlations in the training data and lack sufficient image grounding. To encourage development of models geared towards the latter, we propose a new setting for VQA where for every question type, train and test sets have different prior distributions of answers.

By: Aishwarya Agrawal, Dhruv Batra, Devi Parikh, Aniruddha Kembhavi
June 18, 2018

Detail-Preserving Pooling in Deep Networks

Computer Vision and Pattern Recognition (CVPR)

In this paper, we aim to leverage recent results on image downscaling for the purposes of deep learning.

By: Faraz Saeedan, Nicolas Weber, Michael Goesele, Stefan Roth
June 18, 2018

Deep Spatio-Temporal Random Fields for Efficient Video Segmentation

Computer Vision and Pattern Recognition (CVPR)

In this work we introduce a time- and memory-efficient method for structured prediction that couples neuron decisions across both space at time. We show that we are able to perform exact and efficient inference on a densely connected spatio-temporal graph by capitalizing on recent advances on deep Gaussian random fields.

By: Siddhartha Chandra, Camille Couprie, Iasonas Kokkinos