Design space exploration (DSE) through system-level simulation is essential for designing energy-efficient asynchronous neuromorphic hardware, which is increasingly promising in edge AI applications. However, there are significant mismatches between system-level predictions and gate-level simulations, resulting in low precision when predicting performance during the DSE process for asynchronous neuromorphic hardware. To address this issue, we put forward ANGraph, a graph neural network (GNN)-based performance prediction framework for asynchronous neuromorphic hardware. In the ANGraph framework, we transform the intermediate representation of systemlevel simulations into graphs, collect gate-level circuit simulation results to build benchmarks with over one million samples, and train a GNN model to predict hardware latency for asynchronous neuromorphic hardware. Additionally, we use a residual network (ResNet)-based method to predict the power consumption of asynchronous neuromorphic hardware. We evaluate these two models on additional datasets without extra training across different scales, process nodes, and traffic patterns of input data. Compared to the latency predictions from the state-of-the-art simulator, we improve the R -square score by 0.69 and reduce root mean square error (RMSE) by 76% on average across all datasets. We also achieve an R-square score of 0.98 and a mean absolute percentage error (MAPE) of 0.88% for the power consumption prediction task. The benchmarks and models are available at https://github.com/HuaGuaiGuai/ANGraph.

The success of AI assistants based on language models (LLMs) hinges crucially on Reinforcement Learning from Human Feedback (RLHF), which enables the generation of responses more aligned with human preferences. As universal AI assistants, there's a growing expectation for them to perform consistently across various domains. However, previous work shows that Reinforcement Learning (RL) often exploits shortcuts to attain high rewards and overlooks challenging samples. This focus on quick reward gains undermines both the stability in training and the model's ability to generalize to new, unseen data. In this work, we propose a novel approach that can learn a consistent policy via RL across various data groups or domains. Given the challenges associated with acquiring group annotations, our method automatically classifies data into different groups, deliberately maximizing performance variance. Then, we optimize the policy to perform well on challenging groups. Lastly, leveraging the established groups, our approach adaptively adjusts the exploration space, allocating more learning capacity to more challenging data and preventing the model from over-optimizing on simpler data. Experimental results indicate that our approach significantly enhances training stability and model generalization.

The manufacturing domain relies on Digital Twins (DTs) to mirror physical systems digitally, facilitating simulation, monitoring, and optimization. However, existing DTs may fail to capture the rich contextual knowledge essential for decision-making in complex manufacturing processes. The evolution to knowledge-enhanced DTs is essential, as it integrates domain-specific knowledge models, enabling a profound understanding of processes. To address this gap, this research introduces a knowledge-enhanced DT framework for the production process. This framework utilizes the ontology-based approach to aid the knowledge integration with the manufacturing DTs. The designed framework consists of three essential layers: The source layer, the Streaming data and knowledge coupling layer, and the Service layer. The proposed framework was further implemented in a lab-scale manufacturing setting and validated through several tests. The results demonstrated the seamless integration of knowledge and streaming data in the production process.