CIPARLABS at IJCCI 2026 – Six Research Contributions on Computational Intelligence, Smart Energy, and Time Series

 The CIPARLABS research group from the Department of Information Engineering, Electronics and Telecommunications (DIET) at Sapienza University of Rome participated in the 18th International Joint Conference on Computational Intelligence (IJCCI 2026), held from October 28 to 30 in Angers, France.

The group presented six scientific contributions, spanning the Neural Computation Theory and Applications (NCTA) and Explainable Artificial Intelligence (EXPLAINS) conferences. The papers reflect ongoing research on intelligent energy management, Renewable Energy Communities, reinforcement learning, time-series analysis and cybersecurity.

Although addressing different application domains, these studies share a common research perspective, combining Computational Intelligence methodologies with the challenges of developing efficient, robust and interpretable intelligent systems. Particular attention is devoted to how learning algorithms and optimization techniques can be evaluated in relation to the physical, operational and informational constraints of real-world applications.

Intelligent Energy Management and Renewable Energy Communities

A significant part of the research presented at IJCCI 2026 concerns the intelligent management of distributed energy systems, with particular emphasis on equitable resource allocation, battery degradation and explainable decision-making.

Game-Theoretic Fair Allocation in MPC-Based Management of RECs

Danial Zendehdel, Enrico De Santis, Antonello Rizzi

How can energy be allocated fairly among members of a Renewable Energy Community whose consumption patterns, generation capacities and storage resources differ?

This paper addresses the problem through a decentralized framework combining Model Predictive Control (MPC) with a cooperative allocation mechanism inspired by Nash bargaining theory. The proposed approach coordinates peer-to-peer energy exchanges according to the participants' energy needs and the costs they would incur without cooperation.

The study introduces a distinction between equality and equity in energy distribution, emphasizing the importance of allocating shared resources according to actual demand. Experiments using real household data demonstrate a 50% improvement in the need-based equity metric compared with a centralized Fair-MPC benchmark, while achieving comparable operational efficiency with approximately 75% less peer-to-peer energy trading.

These findings suggest that fairness-aware coordination can improve the distribution of renewable resources without requiring centralized control of individual energy systems.

A Reinforcement Learning Model for Microgrid Energy Management under Uncertainty and Battery Degradation

Gianluca Ferro, Alessio Orlandi, Mohammad Ali Sherafat, Enrico De Santis, Antonello Rizzi

This contribution investigates how reinforcement learning can support energy management decisions while accounting for the progressive degradation of battery storage systems.

The proposed controller employs Proximal Policy Optimization (PPO) to learn battery charging and discharging strategies that balance electricity costs against the economic consequences of battery wear. Forecast information on photovoltaic generation and electricity demand is incorporated into the decision-making process.

Through a multi-objective Pareto analysis, the learned policies are compared with rule-based control and Model Predictive Control, including an oracle MPC benchmark with perfect future information.

The experiments show that longer forecasting horizons can improve the trade-off between operating costs and battery preservation. Under selected experimental configurations, the reinforcement learning controller also achieves lower total costs than the corresponding oracle MPC strategies.

The study contributes to the development of degradation-aware intelligent control, where the long-term condition of energy storage becomes an explicit component of operational optimization.

Explainable EMS Benchmarking for Renewable Energy Communities through Reproducible Multi-Microgrid Simulation

Enrico De Santis, Gianluca Ferro, Antonello Rizzi

Presented at EXPLAINS 2026, this position paper addresses a methodological question in intelligent energy management research. How can heterogeneous Energy Management Systems be compared meaningfully when their performance depends on the operating scenario, available information and evaluation criteria?

The work introduces an explainable and reproducible benchmarking framework based on multi-microgrid simulation, enabling systematic comparisons between interpretable control strategies and supervised learning-based approaches.

The experimental study comprises 1,800 simulations, covering nine controllers, ten operating scenarios, alternative tariff and forecasting configurations, and multiple evaluation criteria.

The results reveal that controller rankings depend substantially on the selected metrics. A strategy that performs well in reducing grid imports may exhibit different behavior in terms of battery depletion or resilience, making aggregate performance scores insufficient to characterize the underlying operational decisions.

The developed platform also provides an educational environment in which students and researchers can investigate energy flows, controller behavior and performance trade-offs through transparent and reproducible experiments.

Time-Series Analysis, Forecasting and Knowledge Discovery

Two further contributions investigate complementary aspects of time-series intelligence, ranging from the discovery of recurring temporal structures to the use of anomaly information for improving predictive models.

Motif Discovery in Time Series – A Survey

Enrico De Santis, Alessio Martino, Vanessa Pennazzi, Antonello Rizzi

Identifying recurring patterns in temporal data is a fundamental problem in time-series analysis, with applications extending from industrial monitoring and energy systems to biological signals and human activity recognition.

This survey provides a systematic overview of time-series motif discovery, tracing its methodological evolution from early symbolic and approximate techniques to scalable exact algorithms and modern representation-learning approaches.

Particular attention is devoted to Matrix Profile-based methods, variable-length and multivariate motif discovery, and emerging techniques based on neural representations.

By examining algorithmic principles, computational trade-offs and available software implementations, the paper provides a reference for researchers and practitioners interested in identifying meaningful temporal regularities across heterogeneous datasets.

Enhancing Prediction Performance in Time Series Forecasting with Anomaly Signals Detected by Hyperdimensional Computing

Sabereh Taghdisi Rastkar, Saeid Jamili, Enrico De Santis, Antonello Rizzi

Irregular events and anomalous observations can substantially affect the reliability of electricity-demand forecasting, especially during periods of high consumption.

This work proposes a forecasting architecture combining Hyperdimensional Computing (HDC) with Long Short-Term Memory (LSTM) neural networks. An HDC-based anomaly detector analyzes recent consumption patterns and provides an additional contextual signal to the forecasting model, allowing it to account for potentially anomalous observations.

Experiments conducted on five years of hourly Brazilian electricity-consumption data, augmented with controlled anomaly injections, demonstrate a 14.9% reduction in forecasting RMSE over the complete test horizon and an 18.5% reduction during peak-demand intervals, compared with an equivalent LSTM model without anomaly information.

The findings indicate that lightweight anomaly representations can enhance the robustness of neural forecasting models, offering a promising direction for intelligent monitoring and predictive management in smart grids.

Neural Time-Series Classification for Cybersecurity

Severity-Aware Sliding Window Classification with Overlap for DDoS Detection

Eugenio Borrini, Enrico De Santis, Antonello Rizzi

The final contribution applies neural time-series classification to the detection of Distributed Denial-of-Service (DDoS) attacks, which threaten network availability by overwhelming communication or computational resources.

The proposed framework employs LSTM neural networks to analyze overlapping windows of network packets, introducing a severity-aware classification mechanism that considers the proportion of malicious traffic within each window.

A central methodological contribution is the adoption of soft labeling, which represents the composition of network traffic more gradually than conventional hard classification labels, improving the learning process when benign and malicious packets are interleaved.

Experiments on the CICDDoS2019 dataset demonstrate substantial improvements in detection performance. In the binary evaluation combining different attack types, classification accuracy increases from 84.8% to 99.1% when soft labeling replaces the conventional hard-labeling strategy.

The work highlights the benefits of severity-aware temporal representations for developing more reliable network intrusion detection systems.

Looking Ahead

The six contributions presented at IJCCI 2026 illustrate the breadth of CIPARLABS' research activities and the methodological connections between apparently distant application domains.

From cooperative energy management and degradation-aware reinforcement learning to temporal pattern discovery and neural cybersecurity, these studies address the relationship between intelligent decision-making, computational modeling and the interpretation of experimental evidence.

Across these research directions, reproducibility, explainability and context-aware performance evaluation emerge as recurring methodological priorities, guiding the development of Computational Intelligence techniques toward increasingly demanding scientific and technological applications.

Beyond the scientific presentations, IJCCI 2026 provided an opportunity to exchange ideas, discuss emerging research directions and strengthen connections within the international Computational Intelligence community.

We thank our co-authors, collaborators and fellow researchers for the stimulating scientific discussions and shared experiences during the conference.

See you at the next IJCCI!

---

Conference information

IJCCI 2026 – Official Technical Program

CIPARLABS – Sapienza University of Rome

CIPARLABS Team at IEEE WCCI 2026

 

CIPARLABS took part in IEEE WCCI 2026 with a broad set of scientific contributions and organizational activities, reflecting the group’s research agenda across Computational Intelligence, machine learning, complex systems, smart grids, bioinformatics, and sustainable energy management.

Within the Special Session «Scientific ML and Bio-Physical Sensing», the group presented the paper «Decoding Functional Multiplicity: Graph Learning Approaches to Multifunctional Proteins». The work addresses the prediction of protein multifunctionality from three-dimensional molecular structures. Proteins are represented as residue contact networks, enabling the use of graph-based learning methods, including simplicial-complex embeddings, graph kernels, and Graph Neural Networks. The study frames the problem as a multi-label classification task over first-level Enzyme Commission classes and shows that topological structural representations can capture meaningful signals related to the functional multiplicity of proteins.

Two further contributions were presented in the Special Session «Smart Energy, Grid, and Infrastructure», including IJCNN SS35 «Computational Intelligence Techniques for Observable Smart Grid and Sustainable Energy Systems».

The first paper, «Forward–Forward Learning for Imbalanced Tabular Predictive Maintenance on a Real-World Smart-Grid Fault Dataset», investigates a stabilized Forward–Forward learning formulation for predictive maintenance in medium-voltage smart grids. The study evaluates layer-local learning on an imbalanced real-world fault dataset from the Rome power distribution grid, comparing it with standard tabular baselines and back-propagation-based MLPs. The results show that Forward–Forward learning is competitive in terms of PR-AUC and F1-score, while Random Forests remain the strongest overall baseline in the considered setting. The work also analyzes calibration, goodness-margin dynamics, stabilization mechanisms, and timing performance, offering a detailed view of the practical viability of Forward–Forward learning for smart-grid fault detection.

The second paper, «Simulation of Microgrid Energy Management under Battery Degradation Costs: a PPO-Based Reinforcement Learning Approach», focuses on residential microgrid energy management with photovoltaic generation and battery storage. The study introduces a degradation-aware simulation framework in which the battery system is modeled through a Battery Management System based on an equivalent circuit model with State-of-Health-dependent parameters. A Proximal Policy Optimization controller is trained to determine battery charge and discharge setpoints and is compared with a rule-based controller and an oracle Model Predictive Control benchmark. The results show that the learned policies outperform the rule-based baseline and approach the oracle MPC behavior in medium battery-utilization regimes, balancing short-term energy costs with long-term battery degradation.

Alongside the scientific presentations, CIPARLABS also contributed to the organization of IEEE WCCI 2026 through two Special Sessions.

The Special Session «AI for Energy and Resource Analytics», chaired by Enrico De Santis, included the session «Computational Intelligence and AI Applications for Sustainable Energy Management in Smart Grids and Energy Communities», CISEM. The session was organized by Enrico De Santis, Antonello Rizzi, and Danial Zendehdel from the Department of Information Engineering, Electronics and Telecommunications at Sapienza University of Rome.

Session page
https://sites.google.com/uniroma1.it/wcci-ijcnn-cisem2026/home-page

A second Special Session, «AICS: Computational Intelligence for Complex Systems», was organized by Alessio Martino, from the Department of AI, Data and Decision Sciences at LUISS University, together with Enrico De Santis and Antonello Rizzi, from the Department of Information Engineering, Electronics and Telecommunications at Sapienza University of Rome. The session gathered contributions devoted to the use of Computational Intelligence for modeling, analyzing, and interpreting complex systems.

Session page
https://sites.google.com/uniroma1.it/wcci-ijcnn-aics2026

Overall, the participation of CIPARLABS at IEEE WCCI 2026 highlights the group’s interdisciplinary research activity, ranging from graph learning for biological systems to predictive maintenance in power infrastructures and reinforcement learning for intelligent energy management. These contributions document the group’s commitment to developing advanced Computational Intelligence methods for scientific, technological, and industrial problems with significant societal and environmental impact.

Conference Proceedings can be downloade here.

See you at IJCNN 2027 in South Africa!

The CIPARLABS team participates in the annual Open Diet 2026 event at the Faculty of Engineering, Department of Information Engineering, Electronics and Telecommunications (DIET), University of Rome "La Sapienza"

 

The CIPARLABS team is participating in the annual Open Diet 2026 event. This event is held at the Faculty of Engineering, Department of Information Engineering, Electronics and Telecommunications (DIET), University of Rome "La Sapienza".

It's an opportunity to meet high school students or undergraduates and showcase both the educational offerings and the research conducted in the laboratories by the various research teams. CIPARLABS strongly believes in this event because, in addition to disseminating knowledge, it also provides a team-building opportunity and a meeting point for all the DIET staff participating.

Our stand presents the Pattern Recognition and Computational Intelligence courses offered for the Master's Degree in Electronic Engineering and the Master's Degree in Telecommunications Engineering.

We presented a Renewable Energy Community simulator with a graphical interface that provides a simple understanding of how Artificial Intelligence and optimization techniques can be used to manage energy flows, piloting renewable sources and energy storage. The tool was developed for educational purposes.

See you next year at Open Diet 2027!

Beyond Perplexity: A Multi-Faceted Analysis of a Novel Densely Connected Transformer

We are pleased to announce the publication of our new paper, Beyond Perplexity: A Multi-Faceted Analysis of a Novel Densely Connected Transformer. This study examines a central question in contemporary language modeling research. If a Transformer decoder is redesigned so that each layer can access the representations produced by all previous layers, can this richer internal connectivity improve performance in a meaningful way? 

The idea is appealing for a clear reason. In several areas of deep learning, dense connectivity has been associated with improved feature reuse, shorter information paths, and potentially more favorable optimization behavior. In our paper, we test whether this intuition also holds for decoder-only autoregressive Transformers, the architectural family underlying many modern large language models.

To address this question rigorously, we designed a methodology aimed at isolating the effect of connectivity from other factors that often confound architectural comparisons. We compared a standard baseline Transformer decoder with a densely connected decoder on two well-known benchmarks for language modeling, Penn Treebank and WikiText-2. The comparison was carried out under two controlled fairness regimes. In the first, both model families were evaluated under the same training recipe, with shared optimization settings and learning-rate search. In the second, the comparison was constrained by the same parameter budget, so that the dense model could not exceed the baseline in parameter count. This distinction was important because it allowed us to separate the possible effect of dense historical connectivity from the simpler effect of adding more capacity.

The experimental setup was complemented by a precise implementation choice. Both datasets were processed with word-level tokenization, using official train, validation, and test splits, and all perplexity values were computed consistently within that vocabulary space. The two architectures shared the same general decoder-only scaffold, while differing in their internal organization. The baseline followed the standard residual Transformer design, whereas the dense variant introduced concatenation-based historical connections followed by learned projection, allowing each layer to reuse information from earlier layers more directly. 

The paper was guided by three main research questions 

1. Does dense historical connectivity improve test perplexity compared with a standard Transformer decoder under controlled comparison regimes?

2. Which architectural factors matter most within the explored design space, including model width, feed-forward size, depth, and number of attention heads?

3. Do dense and baseline models generate texts with different long-range structural signatures, even when standard predictive metrics do not show a clear advantage?

The answer that emerges is both interesting and methodologically instructive. Dense connectivity does not lead to a systematic reduction in perplexity. On WikiText-2, the baseline remains stronger in both fairness regimes. On Penn Treebank, the gains of the dense model are limited and depend on the comparison setting. This matters because it shows that an architectural idea may be theoretically plausible and still fail to deliver a robust practical advantage once tested under controlled conditions.

A particularly relevant result comes from the ablation study. Within the dense family, the most reliable improvements are associated with depth and feed-forward capacity, rather than with dense connectivity alone. This suggests that much of the observed performance variation is better explained by how model capacity is allocated than by the presence of cross-layer concatenation in itself. In other words, the study helps clarify that dense connectivity interacts with more fundamental architectural factors rather than replacing them as the main driver of performance.

Another important aspect of the work lies in the decision to evaluate the models beyond perplexity. Perplexity remains the standard metric for next-token prediction, but it does not capture every relevant aspect of generated language. For this reason, the paper also includes analyses of learning dynamics, attention behavior, targeted probes, and long-form text generation. The probing tasks and attention diagnostics do not reveal a clear linguistic advantage for the dense architecture in the explored setting, although they do highlight behavioral differences between the two model families.

One of the most original contributions of the paper is the use of Zipf–RQA for analyzing generated text. This framework combines Zipf-rank encoding with Recurrence Quantification Analysis in order to study long-range structural regularities in long-form outputs. Here, the results become especially interesting. Even when perplexity does not improve, the dense and baseline models show systematic structural differences in the organization of generated text. This suggests that architectural changes may alter the global form of language generation even when they do not produce better scores on standard predictive metrics.

From a broader perspective, this is the main message of the article. Evaluating a language model through a single headline number is rarely sufficient for understanding what an architecture is actually doing. A richer methodology, one that combines predictive performance, internal diagnostics, and structural analysis of generated text, can reveal differences that would otherwise remain invisible.

Overall, this publication offers a controlled and transparent contribution to Transformer research. Rather than presenting densification as a simple improvement, it shows where its limits emerge, which design factors matter most, and why multi-faceted evaluation is necessary for understanding architectural innovation in language models. For our lab, this work reflects a broader research direction devoted to studying neural language models as complex systems, whose behavior deserves to be analyzed from several complementary viewpoints.

Please cite as:

De Santis, E., Martino, A., & Rizzi, A. (2026). Beyond Perplexity: A Multi-Faceted Analysis of a Novel Densely Connected Transformer. Applied Sciences, 16(6), 2721. https://doi.org/10.3390/app16062721

BibTeX:

@article{deSantis2026BeyondPerplexity,
author = {De Santis, Enrico and Martino, Alessio and Rizzi, Antonello},
title = {Beyond Perplexity: A Multi-Faceted Analysis of a Novel Densely Connected Transformer},
journal = {Applied Sciences},
year = {2026},
volume = {16},
number = {6},
pages = {2721},
doi = {10.3390/app16062721},
url = {https://doi.org/10.3390/app16062721}
}

Degradation-Aware Reinforcement Learning for Smarter Energy Management at IJCCI 2025, Marbella (Spain)

 

Towards Sustainable and Intelligent Microgrids

At the International Joint Conference on Computational Intelligence (IJCCI 2025), held in Marbella (Spain) from 22 to 24 October, CIPARLABS presented new research on degradation-aware energy management for residential microgrids. The paper, authored by Danial Zendehdel, Gianluca Ferro, Enrico De Santis, and Antonello Rizzi, introduces a reinforcement learning framework that intelligently manages battery usage to balance economic efficiency and battery longevity.

Residential microgrids and Renewable Energy Communities are redefining the energy landscape by promoting decentralized and cooperative energy production. Within these systems, lithium-ion batteries play a crucial role in storing excess solar power and ensuring supply continuity. Yet, their performance is constrained by degradation processes that reduce both capacity and lifespan.

Traditional control strategies often neglect this degradation, focusing only on short-term cost optimization. The work presented by the CIPARLABS team moves beyond that limitation by embedding battery health awareness directly into the control policy.

Reinforcement Learning Meets Battery Physics

The study proposes a Reinforcement Learning (RL) framework based on the Soft Actor-Critic (SAC) algorithm, implemented with Stable Baselines3, to learn optimal energy dispatch policies in real time.

What sets this approach apart is the explicit integration of battery State of Health (SoH) feedback into the agent’s learning loop. The RL agent learns to maximize long-term economic rewards while implicitly minimizing degradation, guided by a simplified but empirically calibrated degradation model derived from NASA’s Li-ion cell datasets.

The microgrid environment modeled in the study includes photovoltaic generation, household consumption profiles, and time-of-use electricity tariffs. The RL agent decides, at every timestep, whether to charge or discharge the battery or exchange energy with the grid. Its performance was benchmarked against a Model Predictive Control (MPC) strategy based on Mixed-Integer Linear Programming.

Learning to Preserve Energy and Battery Life

Simulations were conducted using high-resolution load and solar data from the Pecan Street Inc. Dataport. Over both one-year and ten-year scenarios, the RL-based controller demonstrated remarkable improvements over the MPC baseline.

• For a typical household, the RL agent extended battery life by up to 6.3 % compared to the MPC benchmark.

• It reduced energy purchased from the grid by 45–60 %, while maintaining or improving economic performance.

• Over long-term simulations, the degradation-aware SAC agent lowered total battery wear cost by 6.4 %, reflecting more efficient use of the storage system without compromising availability.

These outcomes reveal that the RL framework not only optimizes daily dispatch decisions but also learns non-linear, context-dependent policies that capture the intricate balance between short-term gain and long-term sustainability.

Implications and Future Work

The results suggest a promising pathway for deploying AI-driven, degradation-aware control systems in residential and community microgrids. Such systems could operate autonomously, adapting to changing conditions and maximizing both user savings and battery lifespan.

The research team plans to extend this work through real-world validation and multi-agent reinforcement learning experiments, where multiple prosumers within a Renewable Energy Community coordinate energy exchanges. The framework is being developed within the MOST – Sustainable Mobility Center and supported by European Union Next-GenerationEU funding under the Italian PNRR program.

Danial Zendehdel at IJCCI 2025

This work continues CIPARLABS’ mission to merge computational intelligence and sustainable energy research, paving the way for smarter, more resilient energy ecosystems.

Cite as:

@inproceedings{Zendehdel2026IJCCI,
  author    = {Danial Zendehdel and Gianluca Ferro and Enrico De Santis and Antonello Rizzi},
  title     = {Degradation-Aware Energy Management in Residential Microgrids: A Reinforcement Learning Framework},
  booktitle = {Proceedings of the 17th International Joint Conference on Computational Intelligence (IJCCI 2025)},
  year      = {2026},
  address   = {Marbella, Spain},
  month     = {October 22--24},
  publisher = {SCITEPRESS -- Science and Technology Publications},
  keywords  = {Reinforcement Learning, Battery Management System, Energy Management, Lithium-ion Batteries, Degradation Modeling, Microgrids},
  note      = {(Presented at IJCCI 2025, Marbella, Spain)},
}

CIPAR Labs at IJCNN 2025: Shaping the Future of AI and Complex Systems in Rome

 

From June 30 to July 5, 2025, our team had the honor of actively contributing to the International Joint Conference on Neural Networks (IJCNN 2025), held this year in the heart of Rome, at the prestigious Pontifical Gregorian University. With nearly 2,000 attendees from all over the world, IJCNN 2025 confirmed its role as one of the premier venues for advancing neural network theory, applications, and interdisciplinary dialogue in AI.

While the Roman summer heat proved intense, the conference was intellectually energizing. Our team from CIPAR Labs (Department of Information Engineering, Electronics and Telecommunications, Sapienza University of Rome) took part in a variety of roles: as organizers, speakers, and authors of innovative research at the intersection of AI, smart grids, and complexity science.

One of the standout moments of IJCNN 2025 was the keynote speech by Samy Bengio, Head of AI and Machine Learning at Apple Inc., and a foundational figure in deep learning research. In his talk, Bengio offered a nuanced and thought-provoking perspective on the current state and limitations of large language models (LLMs). While expressing healthy skepticism about their reasoning capabilities, he also highlighted their immense untapped potential, especially when trained on carefully structured and sufficiently rich datasets. After the keynote, we had the opportunity to engage in a stimulating conversation with him, discussing the future trajectories of neural architectures and the balance between performance and interpretability. It was a meaningful exchange that reaffirmed how even at the frontiers of AI, curiosity and critical thinking remain essential drivers of innovation.

Organizing CISEM 2025: Computational Intelligence for Sustainable Energy Management

We proudly organized the CISEM 2025 Workshop — Computational Intelligence for Sustainable Energy Management in Microgrids and Renewable Energy Communities. The workshop addressed the growing demand for intelligent, interpretable, and efficient energy systems, presenting novel techniques for:

• Smart grid optimization and predictive maintenance

• Time-series forecasting

• Multi-agent systems in decentralized energy infrastructures

• Integration of Explainable AI (XAI) and cybersecurity frameworks

The workshop attracted a multidisciplinary audience and was praised for its balance between theory and real-world applications. Contributions came from both academia and industry, reinforcing our commitment to bridging AI research and sustainable energy innovation.

Thanks to Giulia Tanoni (Università Politecnica delle Marche (UNIVPM)) for her tutorial by the title: "Unlocking the potential of industrial non-intrusive load monitoring and future opportunities".

Co-Chairing the AICS Special Session: Artificial Intelligence and Complex Systems

Another milestone was co-organizing the first edition of the AICS Special Session – Artificial Intelligence and Complex Systems. This track explored deep connections between:

• Complexity theory and neural computation

• Emergence, stochastic dynamics, and agent-based modeling

• Philosophical, linguistic, and cognitive perspectives on AI systems

• Foundations of intelligence and model explainability

The AICS session stood out for its interdisciplinary ambition, offering a space where cognitive science, philosophy, and deep learning could meet. We are excited to evolve this session into a full workshop at next year’s conference.

Tutorial: Predictive Maintenance Meets Industry 5.0

During the DLT-4-OSGE Workshop (Deep Learning Techniques for Observable Smart Grid and Sustainable Energy Systems), Prof. Enrico De Santis delivered the tutorial:

“Machine Learning Techniques for Predictive Maintenance, Energy Efficiency and Industry 5.0 Applications”

The session focused on:

• Practical AI architectures for maintenance in industrial and energy systems

• Use of interpretable models (like, Clustering, KANs and SHAP) for actionable insight

• Applications in smart buildings, plants, and Renewable Energy Communities

The tutorial generated vibrant discussions and established the relevance of XAI in high-stakes energy contexts.

Research Papers: Seven Contributions Across Three Tracks

DLT-4-OSGE Workshop

1. On a Fast and Explainable REC HEMS Based on Kolmogorov-Arnold Networks
   Capillo, De Santis, Rizzi
   KANs optimized with Genetic Algorithms for REC energy management, delivering high accuracy and 10% faster inference with explainable output.

2. Decision Focused Forecasting for Smart Grid Energy Management Systems
   Ferro, De Santis, Capillo, Rizzi
   End-to-end LSTM forecasting via Decision Focused Learning, reducing operational costs by 11% over standard pipelines.

3. Multi-Objective Battery Dispatching using an Enhanced SAC Algorithm
   Zendehdel, De Santis, Capillo, Odonkor, Rizzi
   A Lagrangian-penalized SAC algorithm to optimize battery dispatch, improving self-sufficiency ratio (SSR) by 18.2%.

CISEM 2025 Workshop

4. Graph-Augmented LSTM with Weighted Loss for Enhanced Energy Forecasting
   Taghdisi Rastkar, Jamili, De Santis, Rizzi
   Incorporates a GAT-based feature graph and peak-aware loss for accurate, interpretable load forecasting during critical periods.

5. A KAN–SHAP Framework for Fault Detection and Analysis in Smart Grids
   De Santis, Ferro, Rizzi
   Combines Kolmogorov–Arnold Networks with SHAP to classify grid faults with ROC AUC 0.993, enabling actionable fault interpretation and predictive maintenance in the ACEA MV grid.

AICS Special Session

6. LSTM in Recursive Feedback Loops: A Study on Textual Evolution and Complexity
   De Santis, Martino, Ronci, Rizzi
   Theoretical study of semantic self-organization in recursive LSTM generations, analyzing how complexity and feedback impact text evolution.

7. 2025: A GPT Odyssey. Deconstructing Intelligence by Gradual Dissolution of a Transformer
   De Santis, Martino, Bruno, Rizzi
   A HAL 9000-inspired investigation: sequential ablation of GPT-2 layers reveals semantic and structural roles in generative performance, opening paths to explainability in large language models.

Final Reflections

Beyond the papers and presentations, what made IJCNN 2025 so memorable was the spirit of collaboration: colleagues from Brazil, UAE, Spain, UK, and Italy shared not only research, but laughter, meals, and vision.

In a world often divided, the conference served as a reminder that scientific cooperation transcends borders, and that intelligence – whether artificial or human – grows best through connection.

We’re grateful to all our co-authors, collaborators, organizers, and peers who contributed to this vibrant experience.

See you next year in Maastricht at WCCI 2026!

Solar and Wind Forecasting: Rethinking the Future with a Multi-Site Mindset

 

A review of solar and wind energy forecasting: From single-site to multi-site paradigm

The global energy transition is no longer a distant vision — it’s unfolding now, rapidly, and with it comes a crucial question: how do we predict the unpredictable? When it comes to solar and wind energy, forecasting isn’t just a technical detail; it’s a keystone of modern energy systems. In a new paper just published in Applied Energy, Alessio Verdone, Massimo Panella, Enrico De Santis, and Antonello Rizzi from Sapienza University of Rome offer a timely and in-depth exploration of how forecasting methods have evolved to meet this challenge.

Their work is more than just a literature review. It’s a methodological study around the transformation of forecasting paradigms, tracing the field’s progress from early single-site statistical models to the latest deep learning architectures that analyze spatio-temporal data from entire networks of plants. The authors shed light on how our understanding — and our tools — have shifted alongside the growing complexity of renewable energy infrastructures.

At the heart of the paper is a simple but powerful idea: renewable energy production is no longer a local matter. Today’s systems consist of distributed solar panels and wind farms spread across vast areas. By treating each site in isolation, we miss the chance to capture valuable correlations between them. This is where multi-site forecasting comes into play, allowing models to learn not just from the past of a single plant, but from the coordinated behavior of many. And thanks to innovations in machine learning—particularly Graph Neural Networks, Transformers, and hybrid architectures — we now have the tools to make this possible.

The paper is rich with insights. It offers a structured classification of forecasting methods and benchmarks, highlights the most commonly used datasets (and the difficulty in accessing reliable public data), and discusses the metrics used to evaluate performance. But what makes this work stand out is the authors’ critical perspective. They don’t just describe methods — they ask what works, what doesn’t, and why. Their analysis of how spatial and temporal data can be integrated to boost performance speaks directly to current needs in grid management and renewable energy communities.

For researchers, engineers, and energy planners, this review is a valuable resource. It connects the dots between methodological innovation and practical application, offering a clear picture of where the field stands and where it’s heading. More importantly, it invites readers to think systemically: to see renewable energy forecasting not as a single algorithmic task, but as a complex, multi-layered problem with implications for sustainability, policy, and technology.

If you’re interested in the intersection of AI and energy, or if you’re working on Smart Grids or Renewable Energy Communities, forecasting tools, or the design of future energy systems, this is a paper worth diving into.

Read the full paper here!