In this paper, we develop a theory of learning nonlinear input-output maps with fading memory by dissipative quantum systems, as a quantum counterpart of the theory of approximating such maps using classical dynamical systems. The theory identifies the properties required for a class of dissipative quantum systems to be {\em universal}, in that any input-output map with fading memory can be approximated arbitrarily closely by an element of this class. We then introduce an example class of dissipative quantum systems that is provably universal. Numerical experiments illustrate that with a small number of qubits, this class can achieve comparable performance to classical learning schemes with a large number of tunable parameters. Further numerical analysis suggests that the exponentially increasing Hilbert space presents a potential resource for dissipative quantum systems to surpass classical learning schemes for input-output maps.

We evaluate how modern outlier detection methods perform in identifying outliers in e-commerce conversion rate data. Based on the limitations identified, we then present a novel method to detect outliers in e-commerce conversion rate. This unsupervised method is made more business relevant by letting it automatically adjust the sensitivity based on the activity observed on the e-commerce platform. We call this outlier detection method the fluid IQR. Using real e-commerce conversion data acquired from a known store, we compare the performance of the existing and the new outlier detection methods. Fluid IQR method outperforms the existing outlier detection methods by a large margin when it comes to business-relevance. Furthermore, the fluids IQR method is the most robust outlier detection method in the presence of clusters of extreme outliers or level shifts. Future research will evaluate how the fluid IQR method perform in diverse e-business settings.

We consider an agent who is involved in a Markov decision process and receives a vector of outcomes every round. Her objective is to maximize a global concave reward function on the average vectorial outcome. The problem models applications such as multi-objective optimization, maximum entropy exploration, and constrained optimization in Markovian environments. In our general setting where a stationary policy could have multiple recurrent classes, the agent faces a subtle yet consequential trade-off in alternating among different actions for balancing the vectorial outcomes. In particular, stationary policies are in general sub-optimal. We propose a no-regret algorithm based on online convex optimization (OCO) tools (Agrawal and Devanur 2014) and UCRL2 (Jaksch et al. 2010). Importantly, we introduce a novel gradient threshold procedure, which carefully controls the switches among actions to handle the subtle trade-off. By delaying the gradient updates, our procedure produces a non-stationary policy that diversifies the outcomes for optimizing the objective. The procedure is compatible with a variety of OCO tools.

Recent technological advances have proliferated the available computing power, memory, and speed of modern Central Processing Units (CPUs), Graphics Processing Units (GPUs), and Field Programmable Gate Arrays (FPGAs). Consequently, the performance and complexity of Artificial Neural Networks (ANNs) is burgeoning. While GPU accelerated Deep Neural Networks (DNNs) currently offer state-of-the-art performance, they consume large amounts of power. Training such networks on CPUs is inefficient, as data throughput and parallel computation is limited. FPGAs are considered a suitable candidate for performance critical, low power systems, e.g. the Internet of Things (IOT) edge devices. Using the Xilinx SDAccel or Intel FPGA SDK for OpenCL development environment, networks described using the high-level OpenCL framework can be accelerated on heterogeneous platforms. Moreover, the resource utilization and power consumption of DNNs can be further enhanced by utilizing regularization techniques that binarize network weights. In this paper, we introduce, to the best of our knowledge, the first FPGA-accelerated stochastically binarized DNN implementations, and compare them to implementations accelerated using both GPUs and FPGAs. Our developed networks are trained and benchmarked using the popular MNIST and CIFAR-10 datasets, and achieve near state-of-the-art performance, while offering a >16-fold improvement in power consumption, compared to conventional GPU-accelerated networks. Both our FPGA-accelerated determinsitic and stochastic BNNs reduce inference times on MNIST and CIFAR-10 by >9.89x and >9.91x, respectively.

Penetration testing (pentesting) involves performing a controlled attack on a computer system in order to assess it's security. Although an effective method for testing security, pentesting requires highly skilled practitioners and currently there is a growing shortage of skilled cyber security professionals. One avenue for alleviating this problem is automate the pentesting process using artificial intelligence techniques. Current approaches to automated pentesting have relied on model-based planning, however the cyber security landscape is rapidly changing making maintaining up-to-date models of exploits a challenge. This project investigated the application of model-free Reinforcement Learning (RL) to automated pentesting. Model-free RL has the key advantage over model-based planning of not requiring a model of the environment, instead learning the best policy through interaction with the environment. We first designed and built a fast, low compute simulator for training and testing autonomous pentesting agents. We did this by framing pentesting as a Markov Decision Process with the known configuration of the network as states, the available scans and exploits as actions, the reward determined by the value of machines on the network. We then used this simulator to investigate the application of model-free RL to pentesting. We tested the standard Q-learning algorithm using both tabular and neural network based implementations. We found that within the simulated environment both tabular and neural network implementations were able to find optimal attack paths for a range of different network topologies and sizes without having a model of action behaviour. However, the implemented algorithms were only practical for smaller networks and numbers of actions. Further work is needed in developing scalable RL algorithms and testing these algorithms in larger and higher fidelity environments.

Multi-view Multi-instance Multi-label Learning(M3L) deals with complex objects encompassing diverse instances, represented with different feature views, and annotated with multiple labels. Existing M3L solutions only partially explore the inter or intra relations between objects (or bags), instances, and labels, which can convey important contextual information for M3L. As such, they may have a compromised performance. In this paper, we propose a collaborative matrix factorization based solution called M3Lcmf. M3Lcmf first uses a heterogeneous network composed of nodes of bags, instances, and labels, to encode different types of relations via multiple relational data matrices. To preserve the intrinsic structure of the data matrices, M3Lcmf collaboratively factorizes them into low-rank matrices, explores the latent relationships between bags, instances, and labels, and selectively merges the data matrices. An aggregation scheme is further introduced to aggregate the instance-level labels into bag-level and to guide the factorization. An empirical study on benchmark datasets show that M3Lcmf outperforms other related competitive solutions both in the instance-level and bag-level prediction.

Interaction with games can induce emotional reactions which could have an impact on players’ game experience and performance. Physiological sensors such as EEG and eye tracking represent an important mean to track these emotional reactions. In addition, virtual reality isolates the players from the external environment, strengthening the emotional measures. In this paper, we present an explorative study of the use of eye tracking for game adaptation according to the players’ excitement. Results showed that there exists a relationship between the modification of the game’s speed and the EEG excitement index and a correlation between eye movement and excitement as well. These results suggest that eye tracking could be a valid support or replacement of EEG data in game adaptation.

The increasing use of social networks has made opinion mining an important field in the area of Natural Language Processing. The analysis of texts from the reader perspective tends to generate multi-label data since one can interpret the text using different contexts. In this paper, a new method for multi-label classification is proposed to identify reactions or emotions in texts. The new method uses data correlation to improve the class ensemble process used to create the classifiers. In addition to the new method, a new corpus of news written in Brazilian Portuguese labeled with user reactions is presented. Experiments performed with the new corpus and with two existing corpora have demonstrated that the proposed method generates statistically superior or equivalent results, requiring fewer classifiers or classes than traditional problem transformation methods.

Forgetting as a knowledge management operation has received much less attention than operations like inference, or revision. It was mainly in the area of logic programming that techniques and axiomatic properties have been studied systematically. However, at least from a cognitive view, forgetting plays an important role in restructuring and reorganizing a human's mind, and it is closely related to notions like relevance and independence which are crucial to knowledge representation and reasoning. In this paper, we propose axiomatic properties of (intentional) forgetting for general epistemic frameworks which are inspired by those for logic programming, and we evaluate various forgetting operations which have been proposed recently by Beierle et al. according to them. The general aim of this paper is to advance formal studies of (intentional) forgetting operators while capturing the many facets of forgetting in a unifying framework in which different forgetting operators can be contrasted and distinguished by means of formal properties.

Artificial intelligence can help to bring together the twin megatrends of digitalization and decarbonisation. There has been a lot of talk about how artificial intelligence (AI) will affect various aspects of our lives, but little has been said to date about how the technology can help to make the world more sustainable. A new report from the consultancy PwC, commissioned by software giant Microsoft, looks at how the twin, powerfully disruptive megatrends of digitization and decarbonisation could come together in future and it concludes that AI could make a significant dent in global greenhouse gas (GHG) emissions. PwC defines AI as "a collective term for technologies that can sense their environment, think, learn, and take action in response to what they're sensing and their objectives". Applications can range from automation of routine tasks to augmenting human decision-making and beyond to automation and discovery – huge amounts of data to spot, and act on patterns, which are beyond our current capabilities.