In September of 2023, we had the honor of collaborating with the FAIEMA scientific conference as a community partner. The conference provided an excellent platform for our members to showcase their cutting-edge research and innovations in artificial intelligence! We proudly submitted three papers to the conference, reflecting the diverse expertise and contributions of our community. Here’s a brief overview of the papers:
1. A Low-Power Analog Bell-Shaped Classifier Based on Parallel-Connected Gaussian Function Circuits
Abstract:
In this study, a novel approach is presented for developing ultra-low power analog classifiers capable of effectively handling multiple input features while maintaining high levels of accuracy and minimizing power consumption. The proposed methodology is built upon a Voting model that leverages Gaussian likelihood functions. To evaluate the performance of the proposed methodology, a comparison is conducted against the widely used analog Bell-shaped classifiers. Real-life breast cancer dataset is employed for this comparison. The models are trained and the results are processed using the Python programming language. The hardware design and result processing utilize Cadence IC Suite, implementing the TSMC 90 nm CMOS process technology.
AI Catalyst is truly honored to share that this paper has been awarded for its groundbreaking contributions to the field. This recognition serves as a testament to the dedication and collaborative efforts of the research team, as we aim to push the boundaries of knowledge in AI. Written by Vassilis Alimisis, Argyro Kamperi, Nikolaos P. Eleftheriou and Paul P. Sotiriadis.
More details here!
2. The Machine Learning Pod Canvas (MLPod™ Canvas)
Abstract:
A methodology is introduced to assist the development of a product based on the Machine Learning technology. Augmenting the concept of the Business Model Canvas we propose a similar concept, named MLPod™ Canvas, that takes into consideration both technical and business requirements and assists the core product development team to clearly identify the tasks and the requirements for delivering a functional and compliant product. Hints regarding the completion of the canvas are also provided along with explanations about the purpose of each canvas’ individual block.
Written by Demetrios P. Gerogiannis, Paul P. Sotiriadis and Christophoros Nikou. More details here!
3. One-Day-Ahead Wind Speed Forecasting Based on Advanced Deep and Hybrid Quantum Machine Learning
Abstract:
Electricity demand has been rising significantly over the past few years, making it crucial to integrate renewable energy sources (RES) into power networks on a wide scale. Among the most popular alternative energy sources with very high potential is wind energy. However, there is significant variability in wind speed, which results in significant fluctuations in the electricity production from the wind energy. As a result, it is challenging to integrate RES technology and especially wind energy into electricity networks. More accurate forecasts are necessary for the efficient operation of RES power plants, as well as the provision of high-quality electricity at the most affordable prices and the secure and stable operation of electrical networks. Four deep machine learning (ML) algorithms, i.e. multi-head CNN, hybrid quantum multi-head CNN, multi-channel CNN, and encoder–decoder LSTM were applied in this study to estimate medium-term (24 h ahead) wind speed using a real-time measurement dataset.
Written by Konstantinos Blazakis, Yiannis Katsigiannis, Nikolaos Schetakis and AI Catalyst Member Giorgos Stavrakakis. More info here!
Our members proved to be a prominent force in the field. Thus, we encourage the broader AI community to explore these papers, engage with the content, and celebrate the success of AICatalyst in contributing to the advancement of artificial intelligence!
Closing, we look ahead with great anticipation. AICatalyst remains dedicated to fostering productivity and creativity, pushing the boundaries of thought, and continuing our pursuit of excellence in AI. We eagerly anticipate contributing even more research in the future, leaving a lasting industry wide impact. Stay tuned for our continued journey of exploration in the dynamic world of artificial intelligence!
