Photonic Differential Privacy with Direct Feedback Alignment
- URL: http://arxiv.org/abs/2106.03645v1
- Date: Mon, 7 Jun 2021 14:18:01 GMT
- Title: Photonic Differential Privacy with Direct Feedback Alignment
- Authors: Ruben Ohana, Hamlet J. Medina Ruiz, Julien Launay, Alessandro
Cappelli, Iacopo Poli, Liva Ralaivola, Alain Rakotomamonjy
- Abstract summary: We show how to leverage the intrinsic noise of optical random projections to build a differentially private DFA mechanism.
We conduct experiments demonstrating the ability of our learning procedure to achieve solid end-task performance.
- Score: 66.61196212740359
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Optical Processing Units (OPUs) -- low-power photonic chips dedicated to
large scale random projections -- have been used in previous work to train deep
neural networks using Direct Feedback Alignment (DFA), an effective alternative
to backpropagation. Here, we demonstrate how to leverage the intrinsic noise of
optical random projections to build a differentially private DFA mechanism,
making OPUs a solution of choice to provide a private-by-design training. We
provide a theoretical analysis of our adaptive privacy mechanism, carefully
measuring how the noise of optical random projections propagates in the process
and gives rise to provable Differential Privacy. Finally, we conduct
experiments demonstrating the ability of our learning procedure to achieve
solid end-task performance.
Related papers
- Deep Generative Neural Embeddings for High Dimensional Data
Visualization [0.45687771576879593]
We propose a visualization technique that utilizes neural network embeddings and a generative network to reconstruct original data.
We have evaluated the effectiveness of this technique in data visualization and compared it to t-SNE and VAE methods.
Our technique has potential applications in human-in-the-loop training, as it allows for independent editing of embedding locations without affecting the optimization process.
arXiv Detail & Related papers (2023-01-25T14:18:09Z) - On the Importance of Architecture and Feature Selection in
Differentially Private Machine Learning [2.387686431425822]
We study a pitfall in the typical workflow for differentially private machine learning.
The use of differentially private learning algorithms in a "drop-in" fashion yields overly complex and poorly performing models.
arXiv Detail & Related papers (2022-05-13T15:57:34Z) - Image-specific Convolutional Kernel Modulation for Single Image
Super-resolution [85.09413241502209]
In this issue, we propose a novel image-specific convolutional modulation kernel (IKM)
We exploit the global contextual information of image or feature to generate an attention weight for adaptively modulating the convolutional kernels.
Experiments on single image super-resolution show that the proposed methods achieve superior performances over state-of-the-art methods.
arXiv Detail & Related papers (2021-11-16T11:05:10Z) - NerfingMVS: Guided Optimization of Neural Radiance Fields for Indoor
Multi-view Stereo [97.07453889070574]
We present a new multi-view depth estimation method that utilizes both conventional SfM reconstruction and learning-based priors.
We show that our proposed framework significantly outperforms state-of-the-art methods on indoor scenes.
arXiv Detail & Related papers (2021-09-02T17:54:31Z) - Deep Reparametrization of Multi-Frame Super-Resolution and Denoising [167.42453826365434]
We propose a deep reparametrization of the maximum a posteriori formulation commonly employed in multi-frame image restoration tasks.
Our approach is derived by introducing a learned error metric and a latent representation of the target image.
We validate our approach through comprehensive experiments on burst denoising and burst super-resolution datasets.
arXiv Detail & Related papers (2021-08-18T17:57:02Z) - NeuralDP Differentially private neural networks by design [61.675604648670095]
We propose NeuralDP, a technique for privatising activations of some layer within a neural network.
We experimentally demonstrate on two datasets that our method offers substantially improved privacy-utility trade-offs compared to DP-SGD.
arXiv Detail & Related papers (2021-07-30T12:40:19Z) - Plug-and-Play external and internal priors for image restoration [0.0]
We propose a new algorithm for image restoration based on a deep denoiser acting on the image.
We prove the effectiveness of the proposed method in restoring noisy images, both in simulated real medical settings.
arXiv Detail & Related papers (2021-02-15T12:19:28Z) - Bayesian optimization for automatic design of face stimuli [2.572404739180802]
We propose a novel framework which combines generative networks (GANs) with Bayesian optimization to identify individual response patterns to many different faces.
Formally, we employ Bayesian optimization to efficiently search the latent space of state-of-the-art GAN models, with the aim to automatically generate novel faces.
We show how the algorithm can efficiently locate an individual's optimal face while mapping out their response across different semantic transformations of a face.
arXiv Detail & Related papers (2020-07-20T10:27:18Z) - Variational Denoising Network: Toward Blind Noise Modeling and Removal [59.36166491196973]
Blind image denoising is an important yet very challenging problem in computer vision.
We propose a new variational inference method, which integrates both noise estimation and image denoising.
arXiv Detail & Related papers (2019-08-29T15:54:06Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.