This preprint shows that Kerenidis and Prakash’s quantum machine learning (QML) algorithm, one of the strongest candidates for provably exponential speedups in QML, does not in fact give an exponential speedup over classical algorithms. https://arxiv.org/pdf/1807.04271.pdf

# Category: News

## Variational Option Discovery Algorithms

Paper Abstract: “We explore methods for option discovery based on variational inference and make two algorithmic contributions. First: we highlight a tight connection between variational option discovery methods and variational autoencoders, and introduce Variational Autoencoding Learning of Options by Reinforcement (VALOR), a new method derived from the connection. In VALOR, the policy encodes contexts from…

## Mysterious ‘Jumping Gene’

The Mysterious ‘Jumping Gene’ That Appears 500,000 Times in Human DNA https://www.theatlantic.com/science/archive/2018/06/line1-jumping-gene/563354/ Shared from my Google feed

## Neural Ordinary Differential Equations

https://arxiv.org/abs/1806.07366 paper abstract: “We introduce a new family of deep neural network models. Instead of specifying a discrete sequence of hidden layers, we parameterize the derivative of the hidden state using a neural network. The output of the network is computed using a blackbox differential equation solver. These continuous-depth models have constant memory cost, adapt…

## A differetiable Dynamic Programming

Differentiable Dynamic Programming for Structured Prediction and Attention

## AI researchers allege that machine learning is alchemy

We need ideas from physics to shed light on the future AI or to be more accurate machine learning research. http://www.sciencemag.org/news/2018/05/ai-researchers-allege-machine-learning-alchemy

## Hallucinogenic Deep Reinforcement Learning

Teaching a machine to master car racing and fireball avoidance through “World Models”

## Welcome to the Pan-Cancer Atlas

From the analysis of over 11,000 tumors from 33 of the most prevalent forms of cancer, the Pan-Cancer Atlas provides a uniquely comprehensive, in-depth, and interconnected understanding of how, where, and why tumors arise in humans. As a singular and unified point of reference, the Pan-Cancer Atlas is an essential resource for the development of…

## Quantum Correlations Reverse Thermodynamic Arrow of Time

A recent experiment shows how quantum mechanics can make heat flow from a cold body to a hot one, an apparent (though not real) violation of the second law of thermodynamics.

## Do Deep Learning Models Have Too Many Parameters? An Information Theory Viewpoint

Deep learning models often have more parameters than observations, and still perform well. This is sometimes described as a paradox. Found this interesting paper: https://arxiv.org/abs/1802.07044