Dynamics of Learning in Recurrent Feature-Discovery Networks
نویسنده
چکیده
The self-organization of recurrent feature-discovery networks is studied from the perspective of dynamical systems. Bifurcation theory reveals parameter regimes in which multiple equilibria or limit cycles coexist with the equilibrium at which the networks perform principal component analysis.
منابع مشابه
Learning in Linear Feature-Discovery Networks *
We describe the dynamics of learning in unsupervised linear feature-discovery networks that have recurrent lateral connections. Bifurcation theory provides a description of the location of multiple equilibria and limit cycles in the weight-space dynamics.
متن کاملEfficient Short-Term Electricity Load Forecasting Using Recurrent Neural Networks
Short term load forecasting (STLF) plays an important role in the economic and reliable operation ofpower systems. Electric load demand has a complex profile with many multivariable and nonlineardependencies. In this study, recurrent neural network (RNN) architecture is presented for STLF. Theproposed model is capable of forecasting next 24-hour load profile. The main feature in this networkis ...
متن کاملTool-Body Assimilation Model using Neuro-Dynamical System for Acquiring Representation of Tool Function
This paper proposes a model of tool-body assimilation with recurrent neural networks. Though there are existing models for this phenomenon, almost all the model requires the predefined model of the tool shape and/or active sensing process like grasping to recognize the tool. Our model enables a robot to acquire the representation of the motion dynamics of body with multiple tools, and to recogn...
متن کاملOrganizational Learning and Knowledge Spillover in Innovation Networks: Agent-Based Approach (Extending SKIN Framework)
In knowledge-based economy, knowledge has a public good and non-rivalry nature. Firms build their own knowledge stock not only by means of internal R&D and collaboration with partners, but also by means of previously spilled over knowledge of other firms and public research laboratories (such as universities). Firms based on their absorptive capacity, and level of intra-industry and extra-indus...
متن کاملDetecting Overlapping Communities in Social Networks using Deep Learning
In network analysis, a community is typically considered of as a group of nodes with a great density of edges among themselves and a low density of edges relative to other network parts. Detecting a community structure is important in any network analysis task, especially for revealing patterns between specified nodes. There is a variety of approaches presented in the literature for overlapping...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 1990