Advances in Neural Information Processing Systems 7
By G. Tesauro, David S. Touretzky, Todd K. Leen. 1995
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
- Rachid Guerraoui
- El Mahdi El Mhamdi
- Kamalika Chaudhuri
- Anand D. Sarwate
- Tatsuya Harada
- Peter Kairouz
- Reza Shokri
- Ryuji Hamamoto
Citations in the corpus
- Signal Processing and Machine Learning with Differential Privacy (2013, background): “…will not discuss some of the progress made in software systems engineering for differential privacy [25]–[27], algorithms for computing histo grams and contingency tables [28], [29], or the large body of work on privacypreserving data release (references can be found in recent works [18], [30]).”
- When Neurons Fail (2017, background): “…this quantity only requires looking at the topology of the network, while experimentally assessing the robustness of a network requires the costly experiment of looking at all the possible inputs and testing all the possible configurations of the network corresponding to different failure…”
- Unsupervised Brain Abnormality Detection Using High Fidelity Image Reconstruction Networks (2020, background): “Several reasons are often provided for this, e.g., the limited expressiveness of the VAEs inference models, the injected noise, and unsuitable element-wise measures such as the reconstruction squared error (Larsen et al., 2016; Zhao et al., 2019).”
- When Neurons Fail - Technical Report (2016, background): “We show that, in the first case (crash), the bound is on a polynomial value expressed in terms of the number of neurons per layer, the Lipschitz constant of the neuronal activation function, the number of failing neurons, the synaptic weights and the depth of the layer where the crash occurred.”
- Transfer of learning by composing solutions of elemental sequential tasks (2004, methodology): “…learning methods that approximate dynamic programming, such as temporal di erence (Sutton, 1988; Barto et al., 1983) and Q-learning methods (Watkins, 1989; Barto et al., 1990) can be used to directly estimate an optimal policy without building an explicit model of the dynamics of the environment.”