61,216 Hits in 4.2 sec

Spike-based Learning Rules and Stabilization of Persistent Neural Activity

Xiaohui Xie, H. Sebastian Seung
1999 Neural Information Processing Systems  
Such a learning rule acts to stabilize persistent neural activity patterns in recurrent neural networks.  ...  We analyze the conditions under which synaptic learning rules based on action potential timing can be approximated by learning rules based on firing rates.  ...  PERSISTENT ACTIVITY IN A SPIKING AUTAPSE MODEL The preceding arguments about drift velocity were based on approximate rate-based descriptions of learning and network dynamics.  ... 
dblp:conf/nips/XieS99 fatcat:ywsfeamgtveolclparcmn667t4

A Model Drift Detection and Adaptation Framework for 5G Core Networks [article]

Dimitrios Michael Manias, Ali Chouman, Abdallah Shami
2022 arXiv   pre-print
With an increased focus on intelligence and automation through core network functions such as the NWDAF, service providers are tasked with integrating machine learning models and artificial intelligence  ...  The results of this work demonstrate the ability of the drift detection module to accurately characterize a drifted concept as well as the ability of the drift adaptation module to begin the necessary  ...  As seen in this figure, the number of raised drift alarms has been significantly reduced, demonstrating that the drift adaptation module is leading to the model successfully learning under the presence  ... 
arXiv:2209.06852v1 fatcat:u5ckcucypjcy3hqfunui4eauvi

Unsupervised learning for robust working memory [article]

Jintao Gu, Sukbin Lim
2021 bioRxiv   pre-print
Consistent with the findings of previous works, differential plasticity alone was enough to recover a graded-level persistent activity with less sensitivity to learning parameters.  ...  On the other hand, homeostatic plasticity shows a robust recovery of smooth spatial patterns under particular types of synaptic perturbations, such as perturbations in incoming synapses onto the entire  ...  D: Location-258 coded persistent activity under E-I balance.  ... 
doi:10.1101/2021.05.17.444447 fatcat:cxmtheffczdk3etzs2o5jj4fnu

Statistical Mechanics of On-Line Learning Under Concept Drift

Michiel Straat, Fthi Abadi, Christina Göpfert, Barbara Hammer, Michael Biehl
2018 Entropy  
They indicate that LVQ is capable of tracking a classification scheme under drift to a non-trivial extent.  ...  Furthermore, we show that concept drift can cause the persistence of sub-optimal plateau states in gradient based training of layered neural networks for regression.  ...  Figure 4 . 4 Regression under Concept Drift: Learning Curves.  ... 
doi:10.3390/e20100775 pmid:33265863 fatcat:qcspe6g5zrffdijdbrcmopobua

Immediate saccade amplitude disconjugacy induced by unequal images

Zoï Kapoula, Thomas Eggert, Maria Pia Bucci
1995 Vision Research  
For some subjects this disconjugacy persisted even under subsequent monocular viewing.  ...  Saccades Immediate disconjugacy Associative learning Disparity vergence  ...  Persistence of disconjugacy under monocular viewing The persistence of the induced disconjugacy under monocular viewing could be explained in the context of the associative learning hypothesis as follows  ... 
doi:10.1016/0042-6989(95)00150-d pmid:8560815 fatcat:plwp7wiqmjg6nadpu3kxstxwka

Saccade amplitude disconjugacy induced by aniseikonia: role of monocular depth cues

Maria Pia Bucci, Zoı̈ Kapoula, Thomas Eggert
1999 Vision Research  
Disconjugacy persists even in the absence of disparity which indicates learning.  ...  The complex image which had a large variety of monocular depth cues produced the most variable and less persistent disconjugacy.  ...  Both studies reported persistence of the disconjugacy under the subsequently recorded monocular viewing condition. This indicates the presence of a fast learning mechanism.  ... 
doi:10.1016/s0042-6989(99)00064-4 pmid:10664808 fatcat:tgap5bg4h5dung24and744yv7y

Simulations of Cerebellar Motor Learning: Computational Analysis of Plasticity at the Mossy Fiber to Deep Nucleus Synapse

Javier F. Medina, Michael D. Mauk
1999 Journal of Neuroscience  
In contrast, under the more realistic circumstance where the plasticity rule is applied at all times, nucleus and climbing fibercontrolled rules promote spontaneous drift of the strength of synapses during  ...  These results suggest specific constraints for theories of cerebellar motor learning and have general implications regarding the mechanisms that may contribute to the persistence of memories.  ...  Although simulations with Hebbian and climbing fiber-dependent rules could learn in the absence of background inputs, under more realistic conditions their inherent tendency to produce spontaneous drift  ... 
doi:10.1523/jneurosci.19-16-07140.1999 pmid:10436067 fatcat:kaaa5g34srbwxkeot3x4y2cse4

Page 1594 of The Journal of Business Vol. 79, Issue 3 [page]

2006 The Journal of Business  
Finally, in Garcia, Luger, and Re- nault’s (2003) utility-based option pricing model, investors learn about the drift and volatility regime of the joint process describing returns and the stochastic discount  ...  Under their assumptions, the IVS depends on an unobservable latent variable characterizing the regime of the economy.  ... 

Model uncertainty and endogenous volatility

William A. Branch, George W. Evans
2007 Review of economic dynamics (Print)  
We show that there may exist multiple Misspecification Equilibria, a subset of which are stable under least squares learning and dynamic predictor selection.  ...  To isolate the effect of parameter drift versus dual learning our strategy is as follows.  ...  In that paper, there is a unique Misspecification Equilibrium which is stable under learning.  ... 
doi:10.1016/ fatcat:hycd5rtmi5ezbe446eosjiseaq

Plasticity and tuning of the time course of analog persistent firing in a neural integrator

G. Major, R. Baker, E. Aksay, H. S. Seung, D. W. Tank
2004 Proceedings of the National Academy of Sciences of the United States of America  
These complex drift cells often showed a drop in maximum persistent firing rate after training to leak.  ...  Persistent firing could be detuned to instability and leak, respectively, along with fixation behavior.  ...  associated with learning (18, 19) .  ... 
doi:10.1073/pnas.0401992101 pmid:15136747 pmcid:PMC419677 fatcat:gu26jcuaprgzbdyi45jfrxgb3i

Data stream mining: methods and challenges for handling concept drift

Scott Wares, John Isaacs, Eyad Elyan
2019 SN Applied Sciences  
drift detection.  ...  The aim of this research is to portray key challenges faced by algorithmic solutions for stream mining, particularly focusing on the prevalent issue of concept drift.  ...  Fig. 1 1 Types of drifts. Classes represented circles existing algorithms in the Learn++ family include Learn++, Learn++.NC, Learn++. MT, Learn++.NSE, Learn++.NIE and Learn++.CDS.  ... 
doi:10.1007/s42452-019-1433-0 fatcat:nw6d7phfzzex3gsn74izxpltju

Data streams classification by incremental rule learning with parameterized generalization

Francisco Ferrer-Troyano, Jesus S. Aguilar-Ruiz, Jose C. Riquelme
2006 Proceedings of the 2006 ACM symposium on Applied computing - SAC '06  
Mining data streams is a challenging task that requires online systems based on incremental learning approaches.  ...  This paper describes a classification system based on decision rules that may store up-to-date border examples to avoid unnecessary revisions when virtual drifts are present in data.  ...  Virtual drift is consistent but it is not persistent. Noise has neither consistency nor persistence.  ... 
doi:10.1145/1141277.1141428 dblp:conf/sac/Ferrer-TroyanoAS06 fatcat:23re2kq7uzbubdv7okjh6x5o4u

Page 164 of Home Progress Vol. 6, Issue 4 [page]

1916 Home Progress  
He will try this same diving trick, only going deeper, to escape a hawk, plowing along under the drift only to burst up at a dif- ferent spot.  ...  Though in ordinary cold weather he roosts in dense pines and hemlocks, let a snow drift pile up, and he will dive from his tree ‘with a plunge that takes him several inches under the surface and leaves  ... 

Online Learning With Adaptive Rebalancing in Nonstationary Environments

Kleanthis Malialis, Christos G Panayiotou, Marios M Polycarpou
2020 IEEE Transactions on Neural Networks and Learning Systems  
We provide new insights into learning from nonstationary and imbalanced data in online learning, a largely unexplored area.  ...  We compare AREBA with strong baselines and other state-of-the-art algorithms and perform extensive experimental work in scenarios with various class imbalance rates and different concept drift types on  ...  under each type of drift independently.  ... 
doi:10.1109/tnnls.2020.3017863 pmid:32960769 fatcat:cpfx3c6e2zhergyhesd2kskllq

Group selection among alternative evolutionarily stable strategies

Robert Boyd, Peter J. Richerson
1990 Journal of Theoretical Biology  
The rate at which this will occur through drift (or drift-like processes in the case of learning, Cavalli-Sforza & Feldman, 1981) will be slow for sizable groups (Lande, 1985) .  ...  Since smaller groups are likely to have lower persistence, and higher rates of drift, stronger within group processes should lead to more rapid group selection.  ... 
doi:10.1016/s0022-5193(05)80113-4 pmid:2232821 fatcat:t5dfiv7cevbdrbr7bigffrxz2y
« Previous Showing results 1 — 15 out of 61,216 results