No abstract

Semiconductor hookup construction (i.e., constructing process tool piping systems) is critical to semiconductor fabrication plant completion. During the conceptual project phase, it is difficult to conduct an accurate cost estimate due to the great amount of uncertain cost items. This study proposes a new model for estimating semiconductor hookup construction project costs. The developed model, called FALCON‐COST, integrates the component ratios method, fuzzy adaptive learning control network (FALCON), fast messy genetic algorithm (fmGA), and three‐point cost estimation method to systematically deal with a cost‐estimating environment involving limited and uncertain data. In addition, the proposed model improves the current FALCON by devising a new algorithm to conduct building block selection and random gene deletion so that fmGA operations can be implemented in FALCON. The results of 54 case studies demonstrate that the proposed model has estimation accuracy of 83.82%, meaning it is approximately 22.74%, 23.08%, and 21.95% more accurate than the conventional average cost method, component ratios method, and modified FALCON‐COST method, respectively. Providing project managers with reliable cost estimates is essential for effectively controlling project costs.

This work presents a sample efficient and effective value-based method, named SMIX(λ), for reinforcement learning in multi-agent environments (MARL) within the paradigm of centralized training with decentralized execution (CTDE), in which learning a stable and generalizable centralized value function (CVF) is crucial. To achieve this, our method carefully combines different elements, including 1) removing the unrealistic centralized greedy assumption during the learning phase, 2) using the λ-return to balance the trade-off between bias and variance and to deal with the environment's non-Markovian property, and 3) adopting an experience-replay style off-policy training. Interestingly, it is revealed that there exists inherent connection between SMIX(λ) and previous off-policy Q(λ) approach for single-agent learning. Experiments on the StarCraft Multi-Agent Challenge (SMAC) benchmark show that the proposed SMIX(λ) algorithm outperforms several state-of-the-art MARL methods by a large margin, and that it can be used as a general tool to improve the overall performance of a CTDE-type method by enhancing the evaluation quality of its CVF. We open-source our code at: https://github.com/chaovven/SMIX.

Automatic spike sorting is a prerequisite for neuroscience research on multichannel extracellular recordings of neuronal activity. A novel spike sorting framework, combining efficient feature extraction and an unsupervised clustering method, is described here. Wavelet transform (WT) is adopted to extract features from each detected spike, and the Kolmogorov-Smirnov test (KS test) is utilized to select discriminative wavelet coefficients from the extracted features. Next, an unsupervised single linkage clustering method based on grey relational analysis (GSLC) is applied for spike clustering. The GSLC uses the grey relational grade as the similarity measure, instead of the Euclidean distance for distance calculation; the number of clusters is automatically determined by the elbow criterion in the threshold-cumulative distribution. Four simulated data sets with four noise levels and electrophysiological data recorded from the subthalamic nucleus of eight patients with Parkinson's disease during deep brain stimulation surgery are used to evaluate the performance of GSLC. Feature extraction results from the use of WT with the KS test indicate a reduced number of feature coefficients, as well as good noise rejection, despite similar spike waveforms. Accordingly, the use of GSLC for spike sorting achieves high classification accuracy in all simulated data sets. Moreover, J-measure results in the electrophysiological data indicating that the quality of spike sorting is adequate with the use of GSLC.

No abstract