Social dilemmas, where mutual cooperation can lead to high payoffs but participants face incentives to cheat, are ubiquitous in multi-agent interaction. We wish to construct agents that cooperate with pure cooperators, avoid exploitation by pure defectors, and incentivize cooperation from the rest. However, often the actions taken by a partner are (partially) unobserved or the consequences of individual actions are hard to predict. We show that in a large class of games good strategies can be constructed by conditioning one's behavior solely on outcomes (ie. one's past rewards). We call this consequentialist conditional cooperation. We show how to construct such strategies using deep reinforcement learning techniques and demonstrate, both analytically and experimentally, that they are effective in social dilemmas beyond simple matrix games. We also show the limitations of relying purely on consequences and discuss the need for understanding both the consequences of and the intentions behind an action.

It has long been known that a single-layer fully-connected neural network with an i.i.d. prior over its parameters is equivalent to a Gaussian process (GP), in the limit of infinite network width. This correspondence enables exact Bayesian inference for infinite width neural networks on regression tasks by means of evaluating the corresponding GP. Recently, kernel functions which mimic multi-layer random neural networks have been developed, but only outside of a Bayesian framework. As such, previous work has not identified that these kernels can be used as covariance functions for GPs and allow fully Bayesian prediction with a deep neural network. In this work, we derive the exact equivalence between infinitely wide deep networks and GPs. We further develop a computationally efficient pipeline to compute the covariance function for these GPs. We then use the resulting GPs to perform Bayesian inference for wide deep neural networks on MNIST and CIFAR-10. We observe that trained neural network accuracy approaches that of the corresponding GP with increasing layer width, and that the GP uncertainty is strongly correlated with trained network prediction error. We further find that test performance increases as finite-width trained networks are made wider and more similar to a GP, and thus that GP predictions typically outperform those of finite-width networks. Finally we connect the performance of these GPs to the recent theory of signal propagation in random neural networks.

Businesses are naturally interested in detecting anomalies in their internal processes, because these can be indicators for fraud and inefficiencies. Within the domain of business intelligence, classic anomaly detection is not very frequently researched. In this paper, we propose a method, using autoencoders, for detecting and analyzing anomalies occurring in the execution of a business process. Our method does not rely on any prior knowledge about the process and can be trained on a noisy dataset already containing the anomalies. We demonstrate its effectiveness by evaluating it on 700 different datasets and testing its performance against three state-of-the-art anomaly detection methods. This paper is an extension of our previous work from 2016 [30]. Compared to the original publication we have further refined the approach in terms of performance and conducted an elaborate evaluation on more sophisticated datasets including real-life event logs from the Business Process Intelligence Challenges of 2012 and 2017. In our experiments our approach reached an F1 score of 0.87, whereas the best unaltered state-of-the-art approach reached an F1 score of 0.72. Furthermore, our approach can be used to analyze the detected anomalies in terms of which event within one execution of the process causes the anomaly.

Thanks to the growing availability of spoofing databases and rapid advances in using them, systems for detecting voice spoofing attacks are becoming more and more capable, and error rates close to zero are being reached for the ASVspoof2015 database. However, speech synthesis and voice conversion paradigms that are not considered in the ASVspoof2015 database are appearing. Such examples include direct waveform modelling and generative adversarial networks. We also need to investigate the feasibility of training spoofing systems using only low-quality found data. For that purpose, we developed a generative adversarial network-based speech enhancement system that improves the quality of speech data found in publicly available sources. Using the enhanced data, we trained state-of-the-art text-to-speech and voice conversion models and evaluated them in terms of perceptual speech quality and speaker similarity. The results show that the enhancement models significantly improved the SNR of low-quality degraded data found in publicly available sources and that they significantly improved the perceptual cleanliness of the source speech without significantly degrading the naturalness of the voice. However, the results also show limitations when generating speech with the low-quality found data.

An ensemble of neural networks is known to be more robust and accurate than an individual network, however usually with linearly-increased cost in both training and testing. In this work, we propose a two-stage method to learn Sparse Structured Ensembles (SSEs) for neural networks. In the first stage, we run SG-MCMC with group sparse priors to draw an ensemble of samples from the posterior distribution of network parameters. In the second stage, we apply weight-pruning to each sampled network and then perform retraining over the remained connections. In this way of learning SSEs with SG-MCMC and pruning, we not only achieve high prediction accuracy since SG-MCMC enhances exploration of the model-parameter space, but also reduce memory and computation cost significantly in both training and testing of NN ensembles. This is thoroughly evaluated in the experiments of learning SSE ensembles of both FNNs and LSTMs. For example, in LSTM based language modeling (LM), we obtain 21% relative reduction in LM perplexity by learning a SSE of 4 large LSTM models, which has only 30% of model parameters and 70% of computations in total, as compared to the baseline large LSTM LM. To the best of our knowledge, this work represents the first methodology and empirical study of integrating SG-MCMC, group sparse prior and network pruning together for learning NN ensembles. Recently there are increasing interests in using ensembles of Deep Neural Networks (DNNs) (Ju et al. (2017); Huang et al. (2017)), which are known to be more robust and accurate than individual networks. An explanation stems from the fact that learning neural networks is an optimization problem with many local minima (Hansen & Salamon (1990)). Multiple models obtained from applying stochastic optimization, e.g. the widely used Stochastic Gradient Descent (SGD) and its variants, converge to different local minima and tend to make different errors.

Generating video frames that accurately predict future world states is challenging. Existing approaches either fail to capture the full distribution of outcomes, or yield blurry generations, or both. In this paper we introduce an unsupervised video generation model that learns a prior model of uncertainty in a given environment. Video frames are generated by drawing samples from this prior and combining them with a deterministic estimate of the future frame. The approach is simple and easily trained end-to-end on a variety of datasets. Sample generations are both varied and sharp, even many frames into the future, and compare favorably to those from existing approaches.

Imagine you wanted to determine how much solar energy could be generated from adding solar cells to a particular house. This is what Google's Project Sunroof does with Deep Learning. Enter an address and Google uses a Deep Learning framework to estimate how much money you could save in energy costs with solar cells over 20 years (see Figure 1). But let's assume there "might" be an even better way to estimate solar energy savings. For example, you want to use Deep Learning to estimate how much solar energy we could generate with solar panels on the Golden Gate Bridge (that probably wouldn't be a very popular decision in San Francisco).

UK companies are setting the global standard for the next wave of breakthroughs, particularly in technology.In 2016, AlphaGo, a computer programme created by London-based engineers, beat Lee Sedol, a top player of the game. The next wave of research for DeepMind will focus on developing advanced algorithms that can help scientists tackle challenges like curing disease, reducing energy consumption, or building revolutionary materials.

Artificial intelligence has quickly evolved from science fiction to digital assistants such as Alexa and Siri learning about our daily lives. A.I. applications are interpreting MRIs and will soon be operating self-driving cars. In personal finance, many of us interact with chat boxes on bank web sites. And in the investing world, robo-advisors are managing portfolios for retail investors. "But we haven't seen the penetration of AI within institutional finance," said Richard Johnson, vice president of market structure and technology at Greenwich Associates on a recent webinar about the evolution of A.I. and current levels of adoption on Wall Street.

A new algorithm developed by Google and its sibling-company Verily Life Sciences can assess a person's risk of cardiovascular diseases. The AI scans images of a person's retina and accurately predicts the risk of major cardiac events such as heart attack or stroke. The robotic AI uses standard retinal scans to analyze the blood vessels in the retina. The study led by researchers from Google, Verily Life Sciences, and the Stanford School of Medicine used deep-learning algorithms to extract new knowledge from retinal images. The AI scanned the retinal images and identified cardiovascular risk factors based on age, gender, blood pressure, and smoking status, among others.