The purpose of this is to scale down the logits before softmax is applied, a technique similar to the one seen in V aswani et al. (2017). The reason could be that softmax is more numerically stable for both feedforward and backpropagation. As discussed in Section 3.3 of the paper, the stick-breaking formulation was initially used to reflect the process that a Thank you all for your detailed review and insightful comments. This will be the direction in which we take our future work. We have conducted an ablation test for the Gated Recursive Cell and Stick-breaking Attention.

As we state at the beginning of Sec. 2, Theorems 1 and 3 hold for any activation function, including We will clarify this point in the camera-ready version. Figure 1: Histogram of correctly and incorrectly classified pictures shows that trained neural networks are far more likely to misclassify points closer to a classification boundary for both the training and test sets. Results are aggregated across 20 different trained neural networks. We will move the MNIST results to the main paper swapping them with the detailed proofs and modify Sec. We will add in the camera-ready version a discussion on the convergence rate to the Gaussian probability distribution.

Reviewer 1: Please discuss the relevance of the considered classification task for applications in biology.

To answer your question about the baseline, we experimented with two new sample audio generated by the same (Karplus-Strong) algorithm and tested against Alexa. The result is shown in Table.1. The musical audio does not fool Alexa. Thank you again for your constructive feedback! Currently, we are also trying to activate the wake-word using our adversary.

Thank you all reviewers for your in-depth comments. We fixed the typos and abbreviations indicated by reviewer 1 in the text. Previously unexplained terms ( e.g., the offset parameter to the noise) We improved the SI by elaborating on Figure 3, fixing missing sections (4.3) and detailing the DVS Abbreviation SNN (Stochastic Neural Network) was changed to StNN. Klambauer et al. in 2017 indeed introduced Our work is different in terms of objective and results. Regarding Eq. (7), our calculations confirm the derivative w.r.t.

We thank the reviewers for their insightful comments and detailed analysis of our work. Low-rank representation of nonsymmetric DPP kernel: The first term on the right side of Eq. 12 will be singular Regarding the time complexity of the low-rank representation, we see from Eq. 12 that the time complexity required to We will add some text to the camera-ready version of our paper to make this point clear. Learning signed determinantal point processes through the principal minor assignment problem. Learning determinantal point processes by corrective negative sampling.

Thank you for your encouraging comments. Thank you for your thorough and helpful review. We appreciate all of your feedback. This is explained in the "sensor selection" paragraph at the end of the paper and We are glad that you understand and appreciate the significance of Theorem 2. Empirical results/better demonstrations It also suggests that with the "right" constraints put in place, a nonlinear method should do very well. For example, we can try multiple process models on the flu data.