Neural Gas for Sequences
نویسنده
چکیده
For unsupervised sequence processing, standard self organizing maps (SOM) can be naturally extended by recurrent connections and explicit context representations. Known models are the temporal Kohonen map (TKM), recursive SOM, SOM for structured data (SOMSD), and HSOM for sequences (HSOM-S). We discuss and compare the capabilities of exemplary approaches to store different types of sequences. A new efficient model, the Merge-SOM (MSOM), is proposed, combining ideas of TKM and SOMSD, and which is well suited for processing sequences with dynamic multimodal densities.
منابع مشابه
Dynamic Simulation of Distillation Sequences in Dew Pointing Unit of South Pars Gas Refinery
The understanding of the dynamic behavior of distillation columns has received considerable attention because distillation is one of the most widely used unit operations in chemical process industries. This paper reports a dynamic simulation study of the possible distillation columns sequences of Dew pointing unit in the second phase of South Pars Gas Refinery. In this unit, three columns are u...
متن کاملPrediction of methanol loss by hydrocarbon gas phase in hydrate inhibition unit by back propagation neural networks
Gas hydrate often occurs in natural gas pipelines and process equipment at high pressure and low temperature. Methanol as a hydrate inhibitor injects to the potential hydrate systems and then recovers from the gas phase and re-injects to the system. Since methanol loss imposes an extra cost on the gas processing plants, designing a process for its reduction is necessary. In this study, an accur...
متن کاملمعرفی شبکه های عصبی پیمانه ای عمیق با ساختار فضایی-زمانی دوگانه جهت بهبود بازشناسی گفتار پیوسته فارسی
In this article, growable deep modular neural networks for continuous speech recognition are introduced. These networks can be grown to implement the spatio-temporal information of the frame sequences at their input layer as well as their labels at the output layer at the same time. The trained neural network with such double spatio-temporal association structure can learn the phonetic sequence...
متن کاملShort-term and Medium-term Gas Demand Load Forecasting by Neural Networks
The ability of Artificial Neural Network (ANN) for estimating the natural gas demand load for the next day and month of the populated cities has shown to be a real concern. As the most applicable network, the ANN with multi-layer back propagation perceptrons is used to approximate functions. Throughout the current work, the daily effective temperature is determined, and then the weather data w...
متن کاملPredicting the Hydrate Formation Temperature by a New Correlation and Neural Network
Gas hydrates are a costly problem when they plug oil and gas pipelines. The best way to determine the HFT and pressure is to measure these conditions experimentally for every gas system. Since this is not practical in terms of time and money, correlations are the other alternative tools. There are a small number of correlations for specific gravity method to predict the hydrate formation. As th...
متن کاملESTIMATION OF GAS HOLDUP AND INPUT POWER IN FROTH FLOTATION USING ARTIFICIAL NEURAL NETWORK
Multivariable regression and artificial neural network procedures were used to modeling of the input power and gas holdup of flotation. The stepwise nonlinear equations have shown greater accuracy than linear ones where they can predict input power, and gas holdup with the correlation coefficients of 0.79 thereby 0.51 in the linear, and R2=0.88 versus 0.52 in the non linear, respectively. ...
متن کامل