Hierarchical Transfer Learning for Cycle Time Forecasting for Semiconductor Wafer Lot under Different Work in Process Levels

The accurate cycle time (CT) prediction of the wafer fabrication remains a tough task, as the system level of work in process (WIP) is fluctuant. Aiming to construct one unified CT forecasting model under dynamic WIP levels, this paper proposes a transfer learning method for finetuning the predicted neural network hierarchically. First, a two-dimensional (2D) convolutional neural network was constructed to predict the CT under a primary WIP level with the input of spatial-temporal characteristics by reorganizing the input parameters. Then, to predict the CT under another WIP level, a hierarchical optimization transfer learning strategy was designed to finetune the prediction model so as to improve the accuracy of the CT forecasting. The experimental results demonstrated that the hierarchically transfer learning approach outperforms the compared methods in the CT forecasting with the fluctuation of WIP levels.

VDAC—A Primal Perspective

The evolution of the eukaryotic cell from the primal endosymbiotic event involved a complex series of adaptations driven primarily by energy optimization. Transfer of genes from endosymbiont to host and concomitant expansion (by infolding) of the endosymbiont’s chemiosmotic membrane greatly increased output of adenosine triphosphate (ATP) and placed selective pressure on the membrane at the host–endosymbiont interface to sustain the energy advantage. It is hypothesized that critical functions at this interface (metabolite exchange, polypeptide import, barrier integrity to proteins and DNA) were managed by a precursor β-barrel protein (“pβB”) from which the voltage-dependent anion-selective channel (VDAC) descended. VDAC’s role as hub for disparate and increasingly complex processes suggests an adaptability that likely springs from a feature inherited from pβB, retained because of important advantages conferred. It is proposed that this property is the remarkable structural flexibility evidenced in VDAC’s gating mechanism, a possible origin of which is discussed.

Optimal Control of Systems with Distributed Parameters

The article presents outcomes of optimizing a distributed control system for objects with distributed parameters. Variational method was applied while optimization. Transfer functions are obtained for main channels of disturbances and controls. To describe processes in the time domain, a Laplace numerical method of inverting the transform is used. Numerical experiment for heat exchange systems of a counterflow heat exchanger is demonstrated. Various controllers have been tested in automatic control systems. Efficiency of distributed control of aeon of single-circuit and double-circuit control systems is shown. In this case, acceleration curves were obtained at the output of the control object. Distributed control functions are also highlighted. The numerical method provides an iterative calculation process with a fairly small number of iterations