This article is an introductory work towards a larger research framework relative to Scientific Prediction. It is a mixed between science and philosophy of science, therefore we can talk about Experimental Philosophy of Science. As a first result, we introduce a new forecasting method based on image completion, named Forecasting Method by Image Inpainting (FM2I). In fact, time series forecasting is transformed into fully images- and signal-based processing procedures. After transforming a time series data into its corresponding image, the problem of data forecasting becomes essentially a problem of image inpainting problem, i.e., completing missing data in the image. An extensive experimental evaluation is conducted using a large dataset proposed by the well-known M3-competition. Results show that FM2I represents an efficient and robust tool for time series forecasting. It has achieved prominent results in terms of accuracy and outperforms the best M3 forecasting methods.

Due to the high variability of the traffic in the radio access network (RAN), fixed network configurations are not flexible to achieve the optimal performance. Our vendors provide several settings of the eNodeB to optimize the RAN performance, such as media access control scheduler, loading balance, etc. But the detailed mechanisms of the eNodeB configurations are usually very complicated and not disclosed, not to mention the large KPIs space needed to be considered. These make constructing simulator, offline tuning, or rule-based solutions difficult. We aim to build an intelligent controller without strong assumption or domain knowledge about the RAN and can run for 24/7 without supervision. To achieve this goal, we first build a closed-loop control testbed RAN in a lab environment with one eNodeB provided by one of the largest wireless vendors and four smartphones. Next, we build a double Q network agent that is trained with the live feedbacks of the key performance indicators from the RAN. Our work proved the effectiveness of applying deep reinforcement learning to improve network performance in a real RAN network environment.

The "You only look once v4"(YOLOv4) is one type of object detection methods in deep learning. YOLOv4-tiny is proposed based on YOLOv4 to simple the network structure and reduce parameters, which makes it be suitable for developing on the mobile and embedded devices. To improve the real-time of object detection, a fast object detection method is proposed based on YOLOv4-tiny. It firstly uses two ResBlock-D modules in ResNet-D network instead of two CSPBlock modules in Yolov4-tiny, which reduces the computation complexity. Secondly, it designs an auxiliary residual network block to extract more feature information of object to reduce detection error. In the design of auxiliary network, two consecutive 3x3 convolutions are used to obtain 5x5 receptive fields to extract global features, and channel attention and spatial attention are also used to extract more effective information. In the end, it merges the auxiliary network and backbone network to construct the whole network structure of improved YOLOv4-tiny. Simulation results show that the proposed method has faster object detection than YOLOv4-tiny and YOLOv3-tiny, and almost the same mean value of average precision as the YOLOv4-tiny. It is more suitable for real-time object detection.

Thermal management in the hyper-scale cloud data centers is a critical problem. Increased host temperature creates hotspots which significantly increases cooling cost and affects reliability. Accurate prediction of host temperature is crucial for managing the resources effectively. Temperature estimation is a non-trivial problem due to thermal variations in the data center. Existing solutions for temperature estimation are inefficient due to their computational complexity and lack of accurate prediction. However, data-driven machine learning methods for temperature prediction is a promising approach. In this regard, we collect and study data from a private cloud and show the presence of thermal variations. We investigate several machine learning models to accurately predict the host temperature. Specifically, we propose a gradient boosting machine learning model for temperature prediction. The experiment results show that our model accurately predicts the temperature with the average RMSE value of 0.05 or an average prediction error of 2.38 degree Celsius, which is 6 degree Celsius less as compared to an existing theoretical model. In addition, we propose a dynamic scheduling algorithm to minimize the peak temperature of hosts. The results show that our algorithm reduces the peak temperature by 6.5 degree Celsius and consumes 34.5% less energy as compared to the baseline algorithm.

As Gaussian processes are integrated into increasingly complex problem settings, analytic solutions to quantities of interest become scarcer and scarcer. Monte Carlo methods act as a convenient bridge for connecting intractable mathematical expressions with actionable estimates via sampling. Conventional approaches for simulating Gaussian process posteriors view samples as vectors drawn from marginal distributions over process values at a finite number of input location. This distribution-based characterization leads to generative strategies that scale cubically in the size of the desired random vector. These methods are, therefore, prohibitively expensive in cases where high-dimensional vectors - let alone continuous functions - are required. In this work, we investigate a different line of reasoning. Rather than focusing on distributions, we articulate Gaussian conditionals at the level of random variables. We show how this pathwise interpretation of conditioning gives rise to a general family of approximations that lend themselves to fast sampling from Gaussian process posteriors. We analyze these methods, along with the approximation errors they introduce, from first principles. We then complement this theory, by exploring the practical ramifications of pathwise conditioning in a various applied settings.

Decentralized electricity markets are often dominated by a small set of generator companies who control the majority of the capacity. In this paper, we explore the effect of the total controlled electricity capacity by a single, or group, of generator companies can have on the average electricity price. We demonstrate this through the use of ElecSim, a simulation of a country-wide energy market. We develop a strategic agent, representing a generation company, which uses a deep deterministic policy gradient reinforcement learning algorithm to bid in a uniform pricing electricity market. A uniform pricing market is one where all players are paid the highest accepted price. ElecSim is parameterized to the United Kingdom for the year 2018. This work can help inform policy on how to best regulate a market to ensure that the price of electricity remains competitive. We find that capacity has an impact on the average electricity price in a single year. If any single generator company, or a collaborating group of generator companies, control more than ${\sim}$11$\%$ of generation capacity and bid strategically, prices begin to increase by ${\sim}$25$\%$. The value of ${\sim}$25\% and ${\sim}$11\% may vary between market structures and countries. For instance, different load profiles may favour a particular type of generator or a different distribution of generation capacity. Once the capacity controlled by a generator company, which bids strategically, is higher than ${\sim}$35\%, prices increase exponentially. We observe that the use of a market cap of approximately double the average market price has the effect of significantly decreasing this effect and maintaining a competitive market. A fair and competitive electricity market provides value to consumers and enables a more competitive economy through the utilisation of electricity by both industry and consumers.

Long-term forecasting involves predicting a horizon that is far ahead of the last observation. It is a problem of highly practical relevance, for instance for companies in order to decide upon expensive long-term investments. Despite recent progress and success of Gaussian Processes (GPs) based on Spectral Mixture kernels, long-term forecasting remains a challenging problem for these kernels because they decay exponentially at large horizons. This is mainly due their use of a mixture of Gaussians to model spectral densities. The challenges underlying long-term forecasting become evident by investigating the distribution of the Fourier coefficients of (the training part of) the signal, which is non-smooth, heavy-tailed, sparse and skewed. Notably the heavy tail and skewness characteristics of such distribution in spectral domain allow to capture long range covariance of the signal in the time domain. Motivated by these observations, we propose to model spectral densities using a Skewed Laplace Spectral Mixture (SLSM) due to the skewness of its peaks, sparsity, non-smoothness, and heavy tail characteristics. By applying the inverse Fourier Transform to this spectral density we obtain a new GP kernel for long-term forecasting. Results of extensive experiments, including a multivariate time series, show the beneficial effect of the proposed SLSM kernel for long-term extrapolation and robustness to the choice of the number of mixture components.

In the past decade, advances in Artificial Neural Networks (ANNs) have allowed them to perform extremely well for a wide range of tasks. In fact, they have reached human parity when performing image recognition, for example. Unfortunately, the accuracy of these ANNs comes at the expense of a large number of cache and/or memory accesses and compute operations. Spiking Neural Networks (SNNs), a type of neuromorphic, or brain-inspired network, have recently gained significant interest as power-efficient alternatives to ANNs, because they are sparse, accessing very few weights, and typically only use addition operations instead of the more power-intensive multiply-and-accumulate (MAC) operations. The vast majority of neuromorphic hardware designs support rate-encoded SNNs, where the information is encoded in spike rates. Rate-encoded SNNs could be seen as inefficient as an encoding scheme because it involves the transmission of a large number of spikes. A more efficient encoding scheme, Time-To-First-Spike (TTFS) encoding, encodes information in the relative time of arrival of spikes. While TTFS-encoded SNNs are more efficient than rate-encoded SNNs, they have, up to now, performed poorly in terms of accuracy compared to previous methods. Hence, in this work, we aim to overcome the limitations of TTFS-encoded neuromorphic systems. To accomplish this, we propose: (1) a novel optimization algorithm for TTFS-encoded SNNs converted from ANNs and (2) a novel hardware accelerator for TTFS-encoded SNNs, with a scalable and low-power design. Overall, our work in TTFS encoding and training improves the accuracy of SNNs to achieve state-of-the-art results on MNIST MLPs, while reducing power consumption by 1.46$\times$ over the state-of-the-art neuromorphic hardware.

Information technology (IT) students in Denmark have created a software program that can determine the energy consumption and the amount of carbon dioxide generated by the development of deep learning algorithms. According to their estimates, hardware used to train a deep learning algorithm can use worrying amounts of energy from an environmental standpoint. Whether browsing movies suggested by Netflix based on your viewing history, asking your voice assistant a question or interacting with a chatbot on an e-commerce website, all of these everyday online processes rely on deep learning algorithms. However, developing algorithms contributes to digital pollution. And it's precisely this environmental impact that students from the IT department of the University of Copenhagen have sought to quantify, using their Carbontracker software program.

Artificial Intelligence (AI) has the potential to transform how we fight climate change. However, it also increasingly contributes to it: the carbon footprint of AI will grow exponentially over the next decade, and is projected to grow at a CAGR of nearly 44% globally through 2025. The industry is trending towards bigger models (e.g. GPT-3): these require ever-growing datasets, compute budgets, and incur massive energy bills over the model lifecycle. Computational costs of AI models have been doubling every few months, resulting in an estimated 300,000x increase from 2012-2018.