paper-with-me

홈 › Papers

Adding One Neuron Can Eliminate All Bad Local Minima

2018-05-22 · NeurIPS 2018 12 · Shiyu Liang, Ruoyu Sun, Jason D. Lee, R. Srikant

One of the main difficulties in analyzing neural networks is the non-convexity of the loss function which may have many bad local minima. In this paper, we study the landscape of neural networks for binary classification tasks. Under mild assumptions, we prove that after adding one special neuron with a skip connection to the output, or one special neuron per layer, every local minimum is a global minimum.

📄 PDF Abstract BibTeX arXiv:1805.08671

Code (0)

등록된 구현이 없습니다.

Tasks

AllBinary ClassificationGeneral Classification

Similar Papers 제목 키워드 기반

Elimination of All Bad Local Minima in Deep Learning

2019-01-02 · Kenji Kawaguchi, Leslie Pack Kaelbling

In this paper, we theoretically prove that adding one special neuron per output unit eliminates all suboptimal local minima of any deep neural network, for multi-class classification, binary classification, and regressio…

AllBinary ClassificationDeep LearningGeneral Classification+1

Sub-Optimal Local Minima Exist for Neural Networks with Almost All Non-Linear Activations

2019-11-04 · Tian Ding, Dawei Li, Ruoyu Sun

Does over-parameterization eliminate sub-optimal local minima for neural networks? An affirmative answer was given by a classical result in [59] for 1-hidden-layer wide neural networks. A few recent works have extended t…

All

Deep Encryption: Protecting Pre-Trained Neural Networks with Confusion Neurons

2021-09-29 · Mengbiao Zhao, Shixiong Xu, Jianlong Chang, Lingxi Xie 외

Having consumed huge amounts of training data and computational resource, large-scale pre-trained models are often considered key assets of AI service providers. This raises an important problem: how to prevent these mod…

Position

Annihilation of Spurious Minima in Two-Layer ReLU Networks

2022-10-12 · Yossi Arjevani, Michael Field

We study the optimization problem associated with fitting two-layer ReLU neural networks with respect to the squared loss, where labels are generated by a target network. Use is made of the rich symmetry structure to dev…

Vocal Bursts Valence Prediction

Eliminating all bad Local Minima from Loss Landscapes without even adding an Extra Unit

2019-01-12 · Jascha Sohl-Dickstein, Kenji Kawaguchi

Recent work has noted that all bad local minima can be removed from neural network loss landscapes, by adding a single unit with a particular parameterization. We show that the core technique from these papers can be use…

All