Abstract

Abstract
Machine learning has been facing significant challenges over the last years, much of which stem from the new characteristics of machine learning problems, such as learning from streaming data or incorporating human feedback into existing datasets and models. In these dynamic scenarios, data change over time and models must adapt. However, new data do not necessarily mean new patterns. The main goal of this paper is to devise a method to predict a model's performance metrics before it is trained, in order to decide whether it is worth it to train it or not. That is, will the model hold significantly better results than the current one? To address this issue, we propose the use of meta-learning. Specifically, we evaluate two different meta-models, one built for a specific machine learning problem, and another built based on many different problems, meant to be a generic meta-model, applicable to virtually any problem. In this paper, we focus only on the prediction of the root mean square error (RMSE). Results show that it is possible to accurately predict the RMSE of future models, event in streaming scenarios. Moreover, results also show that it is possible to reduce the need for re-training models between 60% and 98%, depending on the problem and on the threshold used.

Abstract
Lock-free data structures are an important tool for the development of concurrent programs as they provide scalability, low latency and avoid deadlocks, livelocks and priority inversion. However, they require some sort of additional support to guarantee memory reclamation. The Optimistic Access (OA) method has most of the desired properties for memory reclamation, but since it allows memory to be accessed after being reclaimed, it is incompatible with the traditional memory management model. This renders it unable to release memory to the memory allocator/operating system, and, as such, it requires a complex memory recycling mechanism. In this paper, we extend the lock-free general purpose memory allocator LRMalloc to support the OA method. By doing so, we are able to simplify the memory reclamation method implementation and also allow memory to be reused by other parts of the same process. We further exploit the virtual memory system provided by the operating system and hardware in order to make it possible to release reclaimed memory to the operating system.

Abstract
This study presents a planning model for utilizing emergency transmission capacity in the power system reserve market with renewable energy sources. To this end, first, the effects of the operation of a transmission line at higher power than rated power are described. The lifetime reduction of transmission lines caused by operation under these conditions is then measured, and finally, the price is determined based on the rate of lifetime reduction. This surplus capacity is then entered into a two-stage model of the energy and reserve market as a function of price offer, while also taking renewable energy sources into account. The numerical results of a 6-bus network indicates that the introduction of renewable energy sources reduced energy costs while increasing reserve market costs due to uncertainty. Despite the emergency capacity, such costs are reduced due to the network's utilization of low-cost resources.

Abstract
The use of machine learning in embedded systems is an interesting topic, especially with the growth in popularity of the Internet of Things (IoT). The capacity of a system, such as a robot, to self-localize, is a fundamental skill for its navigation and decision-making processes. This work focuses on the feasibility of using machine learning in a Raspberry Pi 4 Model B, solving the localization problem using images and fiducial markers (ArUco markers) in the context of the RobotAtFactory 4.0 competition. The approaches were validated using a realistically simulated scenario. Three algorithms were tested, and all were shown to be a good solution for a limited amount of data. Results also show that when the amount of data grows, only Multi-Layer Perception (MLP) is feasible for the embedded application due to the required training time and the resulting size of the model.

Abstract
The future is envisaged to have renewable energy resources replacing conventional sources of energy like fossil fuels. In this direction wind and solar energy is emerging to be a vital source of green energy. Although these resources are a promising aspect in providing clean and cheap electrical energy, one demerit is that it is intermittent and therefore unpredictable. This intermittent nature poses a challenge in maintaining the balance between generation and demand of electrical energy thus adversely affecting the system control. Also, the electrical energy companies involved in selling by participating in the electricity pool market need highly accurate solar and wind energy predictions for maximizing their profit. These issues demand a tool for accurate prediction of generation. This paper proposes machine learning prediction models for wind and solar irradiation. For this, a case study is done considering weather data of Malviya National Institute of Technology in Jaipur used to train the regression models. The best-trained model is tested with unseen data and shown to have reasonably good accuracy in predicting wind speed and solar irradiation. A comparative study of regression model performances is done. It is shown that Gaussian Process Regression-based prediction for solar irradiation and the Support Vector Machine outperforms the trained model for the wind speed predictions. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.