Transfer Learning (TL) is currently the most effective approach for modeling building thermal dynamics when only limited data are available. TL uses a pretrained model that is fine-tuned to a specific target building. However, it remains unclear how to proceed after initial fine-tuning, as more operational measurement data are collected over time. This challenge becomes even more complex when the dynamics of the building change, for example, after a retrofit or a change in occupancy. In Machine Learning literature, Continual Learning (CL) methods are used to update models of changing systems. TL approaches can also address this challenge by reusing the pretrained model at each update step and fine-tuning it with new measurement data. A comprehensive study on how to incorporate new measurement data over time to improve prediction accuracy and address the challenges of concept drifts (changes in dynamics) for building thermal dynamics is still missing. Therefore, this study compares several CL and TL strategies, as well as a model trained from scratch, for thermal dynamics modeling during building operation. The methods are evaluated using 5--7 years of simulated data representative of single-family houses in Central Europe, including scenarios with concept drifts from retrofits and changes in occupancy. We propose a CL strategy (Seasonal Memory Learning) that provides greater accuracy improvements than existing CL and TL methods, while maintaining low computational effort. SML outperformed the benchmark of initial fine-tuning by 28.1\% without concept drifts and 34.9\% with concept drifts.

Improving energy efficiency in residential buildings is critical to combating climate change and reducing greenhouse gas emissions. Retrofitting existing buildings, which contribute a significant share of energy use, is therefore a key priority, especially in regions with outdated building stock. Artificial Intelligence (AI) and Machine Learning (ML) can automate retrofit decision-making and find retrofit strategies. However, their use faces challenges of data availability, model transparency, and compliance with national and EU AI regulations including the AI act, ethics guidelines and the ALTAI. This paper presents a trustworthy-by-design ML-based decision support framework that recommends energy efficiency strategies for residential buildings using minimal user-accessible inputs. The framework merges Conditional Tabular Generative Adversarial Networks (CTGAN) to augment limited and imbalanced data with a neural network-based multi-label classifier that predicts potential combinations of retrofit actions. To support explanation and trustworthiness, an Explainable AI (XAI) layer using SHapley Additive exPlanations (SHAP) clarifies the rationale behind recommendations and guides feature engineering. Two case studies validate performance and generalization: the first leveraging a well-established, large EPC dataset for England and Wales; the second using a small, imbalanced post-retrofit dataset from Latvia (RETROFIT-LAT). Results show that the framework can handle diverse data conditions and improve performance up to 53% compared to the baseline. Overall, the proposed framework provides a feasible, interpretable, and trustworthy AI system for building retrofit decision support through assured performance, usability, and transparency to aid stakeholders in prioritizing effective energy investments and support regulation-compliant, data-driven innovation in sustainable energy transition.

Survey research is essential in energy demand studies for capturing consumer preferences and informing policy decisions. Stated preference (SP) surveys, in particular, analyse how individuals make trade-offs in hypothetical scenarios. However, traditional survey methods are costly, time-consuming, and affected by biases and respondent fatigue. Large language models (LLMs) have emerged as a potential tool to address these challenges by generating human-like textual responses. This study investigates the ability of LLMs to simulate consumer choices in energy-related SP surveys. A series of test scenarios evaluated the simulation performance of LLMs at both individual and aggregated levels, considering factors in the prompt, in-context learning (ICL), chain-of-thought (CoT) reasoning, the comparison between local and cloud-based LLMs, integration with traditional choice models, and potential biases. Results indicate that while LLMs achieve an average accuracy of up to 48%, surpassing random guessing, their performance remains insufficient for practical application. Local and cloud-based LLMs perform similarly in simulation accuracy but exhibit differences in adherence to prompt requirements and susceptibility to social desirability biases. Findings suggest that previous SP choices are the most effective input factor, while longer prompts with varied factor formats may reduce accuracy. Furthermore, the traditional mixed logit choice model outperforms LLMs and provides insights for refining LLM prompts. Despite their limitations, LLMs provide scalability and efficiency advantages, requiring minimal historical data compared to traditional survey methods. Future research should refine prompt structures, further investigate CoT reasoning, and explore fine-tuning techniques to improve LLM-based energy survey simulations.

I know a lot of developers and software engineers who love writing. You can use ChatGPT in order to help you check errors, grammar or juste rate your article before publishing. If you noticed, the paragraph I gave was the one in the headline. For me, it looks efficient and I don't think I need to add more than that. Instead of asking your friends, let's ask ChatGPT: You can always ask it for improvements and help for your articles, copywriting and anything in your mind.

The need for Energy Savings has become increasily foundamental to fight Climate Change. We have been working on a cloud-based RL algorithm that can retrofit existing HVAC controls to obtain substantial results. In the last decade, a new class of controls which relies on Artificial Intelligence have been proposed. In particular, we are going to highlight data-driven controls based on Reinforcement Learning (RL), since they showed from the very beginning promising results as HVAC controls [2]. There are two main ways to upgrade with RL the air conditioning systems: to implement RL on new systems or to retrofit the existing ones.

Sanofi, which has moved purposefully into high technologies to get more from its manufacturing, will lean heavily on that strategy to shrink costs and fatten margins. Using robotics, artificial intelligence and new generation manufacturing should save it half a billion euros in annual costs by 2022. So says Sanofi CFO Jean-Baptiste Chasseloup de Chatillon who was filling in some details of new CEO Paul Hudson's €2 billion cost-savings plan laid out Tuesday during Sanofi's investor conference. "It is a leapfrogging of productivity. It reduces cycle time," Chasseloup de Chatillon said on a webcast of the conference.

Sanofi, which has moved purposefully into high technologies to get more from its manufacturing, will lean heavily on that strategy to shrink costs and fatten margins. Using robotics, artificial intelligence and new generation manufacturing should save it half a billion euros in annual costs by 2022. So says Sanofi CFO Jean-Baptiste Chasseloup de Chatillon who was filling in some details of new CEO Paul Hudson's €2 billion cost-savings plan laid out Tuesday during Sanofi's investor conference. "It is a leapfrogging of productivity. It reduces cycle time," Chasseloup de Chatillon said on a webcast of the conference.

The tower at 4th and Columbia will be the tallest in Seattle, a 1,029-foot, $290 million monument to the city's recent, tech-flavored success. And, if the current plans are approved, the tower will include a quirky twist: four levels of above-grade parking, designed to someday take on new life as apartments and offices. LMN Architects, which designed the project, wants the tower to survive 50 to 100 years. "If that's the case, we do need to make sure--I feel we do have have the responsibility--that if the parking uses do change, we design to be able to adapt to that change," says John Chau, a partner at the firm. The change he's talking about is the coming transformation to a car-free-ish future.

Drive.ai is a Silicon Valley startup working on a kit to retrofit your ride If Drive.ai is a success, your first self-driving car might already be parked in the driveway. The Silicon Valley start-up, founded recently by a team of former Stanford University Artificial Intelligence Lab products, is working on a software kit that can be used to retrofit existing vehicles. "We started Drive.ai because we believe there's a real opportunity to make our roads, our commutes, and our families safer," the company announced in a statement on its blog, citing a statistic that more than one million people die each year worldwide in automobile accidents caused by human error. At its foundation, Drive.ai is looking to use deep learning -- which its founders consider the most effective form of artificial intelligence ever developed -- to key a breakthrough in a field that giant companies such as Google and General Motors have been trying to master for years. "Unlike other forms of AI, which involve programming many sets of rules, a deep learning algorithm learns more like a human brain. You provide examples, tagged and labeled by an expert, and the system starts to learn for itself -- creating its own rules."

The autonomous driving software that powers Tesla vehicles will be getting a huge upgrade once Tesla rolls out the second iteration of its Autopilot software. Not only will the software itself be smarter, but it will be bolstered by enhanced hardware that will come in the form of additional radar units and camera systems all around the car. Tesla's updated Autopilot software is reportedly in beta testing at the moment and will also accompany an updated UI on the dash that will more accurately display other cars on the road along with the direction and angle they are moving in. Other rumored features include the ability to identify and react to both stop signs and traffic lights. All that said, current Tesla owners with Autopilot enabled vehicles may naturally be wondering if they'll soon be left using outdated, and in turn, riskier hardware once Tesla vehicles fitted with the company's latest and greatest hardware begin rolling off the line.