Inversion of multilayer nets

By A. Linden, Jörg Kindermann. 1989

How researchers see this paper

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

In Model Architectures

Citations in the corpus

Semantic Image Inpainting with Deep Generative Models (2016, methodology): “Additionally, back-propagation has also been used to visualize and understand the learned features in a trained network, by “inverting” the network through updating the gradient at the input layer [26, 5, 35, 21].”
Understanding deep image representations by inverting them (2014, methodology): “Algorithms similar to the back-propagation technique developed here were proposed by [16, 18, 21, 36], along with alternative optimisation strategies based on sampling.”
Inverting Visual Representations with Convolutional Networks (2015, methodology): “These include works making use of backpropagation or sampling [15, 16, 18, 27, 9, 25] and, most similar to our approach, other neural networks [2].”
Visualizing Deep Convolutional Neural Networks Using Natural Pre-images (2015, methodology): “1 It is referred to as TV norm in Mahendran and Vedaldi (2015) but for β = 2 this is actually the quadratic norm. Several pre-image methods alternative to energy minimization have been explored as well. Nguyen et al. (2015) used genetic programming to generate images that maximize the response of se”
Adversarial Transformation Networks: Learning to Generate Adversarial Examples (2017, methodology): “Gatys et al. (2015) originally proposed a gradient descent procedure based on “back-driving networks” (Linden & Kindermann, 1989) to modify the inputs of a fully-trained network to ﬁnd a set of inputs that maximize a desired set of outputs and hidden unit activations.”
Inverting Supervised Representations with Autoregressive Neural Density Models (2018, methodology): “Although our approach has similar goals to optimization-based approaches to network inversion (Linden and Kindermann, 1989; Lee and Kil, 1994; Lu et al., 1999; Mahendran and Vedaldi, 2015) it is most closely related to methods which make use of another neural network that is trained to in-vert the…”
Semantic Image Inpainting with Perceptual and Contextual Losses (2016, methodology): “Additionally, back-propagation has also been used to visualize and understand the learned features in a trained network, by “inverting” the network through updating the gradient at the input layer[5, 15, 17, 23].”
One-step neural network inversion with PDF learning and emulation (2005, unknown): “Algorithms have been proposed for this problem that perform gradient descent on the input vector [6] [7], or use genetic algorithms [8] [9], or use a number of other approaches.”
Salient Deconvolutional Networks (2016, methodology): “SIFT [22], BoVW [2,7], as well as early neural networks [23,10,9,12,6,20].”
Neural network explanation using inversion (2007, background): “The inversion of MLP neural networks has been previously shown to be useful in finding a set of input patterns which produce a target output pattern (Linden & Kindermann, 1989).”
Neural Network Inversion in Adversarial Setting via Background Knowledge Alignment (2019, background): “In general, these inversion approaches fall into two categories: optimization-based inversion [21, 30, 37, 39, 43, 44, 72] and, most similar to our approach, training-based inversion [10, 17, 18, 53].”
Inverting Convolutional Networks with Convolutional Networks (2015, methodology): “These include works making use of backpropagation or sampling [12, 13, 14, 15, 16, 17] and, most similar to our approach, other neural networks [18].”
Learning and the Unknown: Surveying Steps toward Open World Recognition (2019, unknown): “Another decades-old rejection approach used in neural networks is to add a “garbage” class (Linden and Kindermann 1989) or “background” class (Chang and Lippmann 1994).”
Adversarial Neural Network Inversion via Auxiliary Knowledge Alignment (2019, methodology): “[7] Linden and Kindermann, “Inversion of multilayer nets,” in International 1989 Joint Conference on Neural Networks, 1989, pp.”