Fetching the paper…
Reading the bibliography…
It has been widely assumed that a neural network cannot be recovered from its outputs, as the network depends on its parameters in a highly nonlinear way.
Receptive field organization of simple cells in cat striate cortex
Heggelund, P · 1981
Earlier work this paper cites.
Reconstructing a neural net from its output
Fefferman, C · 1994
Earlier work this paper cites.
Gain modulation from background synaptic input
Chance, F. S., Abbott, L. F., and Reyes, A. D · 2002
Earlier work this paper cites.
How are complex cell properties adapted to the statistics of natural stimuli?
Kording, K. P., Kayser, C., Einhauser, W., and Konig, P · 2004
Earlier work this paper cites.
Adam: A method for stochastic optimization
Kingma, D. P. and Ba, J · 2014
Earlier work this paper cites.
Delving deep into rectifiers: Surpassing human-level performance on ImageNet classification
He, K., Zhang, X., Ren, S., and Sun, J · 2015
Earlier work this paper cites.
Representation benefits of deep feedforward networks
Telgarsky, M · 2015
Cited alongside, same era.
Deep residual learning for image recognition
He, K., Zhang, X., Ren, S., and Sun, J · 2016
Cited alongside, same era.
Learning neural networks with two nonlinear layers in polynomial time
Goel, S. and Klivans, A · 2017
Cited alongside, same era.
Reliably learning the ReLU in polynomial time
Goel, S., Kanade, V., Klivans, A., and Thaler, J · 2017
Cited alongside, same era.
On the expressive power of deep neural networks
Raghu, M., Poole, B., Kleinberg, J., Ganguli, S., and Sohl-Dickstein, J · 2017
Cited alongside, same era.
How to start training: The effect of initialization and architecture
Learning two-layer neural networks with symmetric inputs
Ge, R., Kuditipudi, R., Li, Z., and Wang, X · 2019
Closest in time.
High-fidelity extraction of neural network models
Jagielski, M., Carlini, N., Berthelot, D., Kurakin, A., and Papernot, N · 2019
Closest in time.
Model reconstruction from model explanations
Milli, S., Schmidt, L., Dragan, A. D., and Hardt, M · 2019
Closest in time.
Towards reverse-engineering black-box neural networks
Oh, S. J., Schiele, B., and Fritz, M · 2019
Closest in time.
Functional vs. parametric equivalence of relu networks
Phuong, M. and Lampert, C. H · 2019
Closest in time.
alphaXiv searches the wider corpus for related work and actual follow-ups.
alphaXiv is searching for related work…
Hanin, B. and Rolnick, D · 2018
Cited alongside, same era.
Complexity of linear regions in deep networks
Hanin, B. and Rolnick, D
Cited in the paper.
Deep ReLU networks have surprisingly few activation patterns
Hanin, B. and Rolnick, D
Cited in the paper.