Search

We present a strategy for a single quantum miner with relatively low hashing power, with the same ramifications as a 51% attack. Bitcoin nodes consider the chain with the highest cumulative proof-of-work to be the valid chain. A quantum miner can manipulate the block timestamps to multiply the difficulty by $c$. The fork-choice rule counts every block with increased difficulty with weight $c$. By using Grover's algorithm, it is only $O(\sqrt c)$ harder for the quantum miner to mine such blocks. By picking a high enough $c$, the single quantum miner can create a competing chain with fewer blocks, but more cumulative proof-of-work. The time required is $O(\frac{1}{r^2})$ epochs, where $r$ is the fraction of the block rewards that the quantum miner would have received if they mined honestly. Most proof-of-work cryptocurrencies, including Bitcoin, are vulnerable to our attack. However, it will likely be impossible to execute in forthcoming years, as it requires an extremely fast and fault-tolerant quantum computer., Comment: Minor revision

This paper investigates the approximation power of three types of random neural networks: (a) infinite width networks, with weights following an arbitrary distribution; (b) finite width networks obtained by subsampling the preceding infinite width networks; (c) finite width networks obtained by starting with standard Gaussian initialization, and then adding a vanishingly small correction to the weights. The primary result is a fully quantified bound on the rate of approximation of general general continuous functions: in all three cases, a function $f$ can be approximated with complexity $\|f\|_1 (d/\delta)^{\mathcal{O}(d)}$, where $\delta$ depends on continuity properties of $f$ and the complexity measure depends on the weight magnitudes and/or cardinalities. Along the way, a variety of ancillary results are developed: an exact construction of Gaussian densities with infinite width networks, an elementary stand-alone proof scheme for approximation via convolutions of radial basis functions, subsampling rates for infinite width networks, and depth separation for corrected networks., Comment: This submission constitutes a poor approach to the problem, and has no scientific purpose. A superior (different) approach (and stronger final result, also treating the NTK) has appeared in arXiv:1910.06956 ; please see that work instead

This paper provides a simple procedure to fit generative networks to target distributions, with the goal of a small Wasserstein distance (or other optimal transport costs). The approach is based on two principles: (a) if the source randomness of the network is a continuous distribution (the "semi-discrete" setting), then the Wasserstein distance is realized by a deterministic optimal transport mapping; (b) given an optimal transport mapping between a generator network and a target distribution, the Wasserstein distance may be decreased via a regression between the generated data and the mapped target points. The procedure here therefore alternates these two steps, forming an optimal transport and regressing against it, gradually adjusting the generator network towards the target distribution. Mathematically, this approach is shown to minimize the Wasserstein distance to both the empirical target distribution, and also its underlying population counterpart. Empirically, good performance is demonstrated on the training and testing sets of the MNIST and Thin-8 data. The paper closes with a discussion of the unsuitability of the Wasserstein distance for certain tasks, as has been identified in prior work [Arora et al., 2017, Huang et al., 2017]., Comment: Appears in ICML 2019

This paper investigates the ability of generative networks to convert their input noise distributions into other distributions. Firstly, we demonstrate a construction that allows ReLU networks to increase the dimensionality of their noise distribution by implementing a "space-filling" function based on iterated tent maps. We show this construction is optimal by analyzing the number of affine pieces in functions computed by multivariate ReLU networks. Secondly, we provide efficient ways (using polylog $(1/\epsilon)$ nodes) for networks to pass between univariate uniform and normal distributions, using a Taylor series approximation and a binary search gadget for computing function inverses. Lastly, we indicate how high dimensional distributions can be efficiently transformed into low dimensional distributions.