?url_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Adc&rft.title=Dynamical+Recurrent+Neural+Networks%3A+Towards+Environmental+Time+Series+Prediction%7D&rft.creator=Aussem%2C+A.&rft.creator=Murtagh%2C+F.&rft.creator=Sarazin%2C+M.&rft.subject=Artificial+Intelligence&rft.subject=Dynamical+Systems&rft.subject=Neural+Nets&rft.subject=Speech&rft.description=Dynamical+Recurrent+Neural+Networks+(DRNN)+(Aussem+1995a)+are+a+class+of+fully+recurrent+networks+obtained+by+modeling+synapses+as+autoregressive+filters.+By+virtue+of+their+internal+dynamic%2C+these+networks+approximate+the+underlying+law+governing+the+time+series+by+a+system+of+nonlinear+difference+equations+of+internal+variables.+They+therefore+provide+history-sensitive+forecasts+without+having+to+be+explicitly+fed+with+external+memory.+The+model+is+trained+by+a+local+and+recursive+error+propagation+algorithm+called+temporal-recurrent-backpropagation.+The+efficiency+of+the+procedure+benefits+from+the+exponential+decay+of+the+gradient+terms+backpropagated+through+the+adjoint+network.+We+assess+the+predictive+ability+of+the+DRNN+model+with+meteorological+and+astronomical+time+series+recorded+around+the+candidate+observation+sites+for+the+future+VLT+telescope.+The+hope+is+that+reliable+environmental+forecasts+provided+with+the+model+will+allow+the+modern+telescopes+to+be+preset%2C+a+few+hours+in+advance%2C+in+the+most+suited+instrumental+mode.+In+this+perspective%2C+the+model+is+first+appraised+on+precipitation+measurements+with+traditional+nonlinear+AR+and+ARMA+techniques+using+feedforward+networks.+Then+we+tackle+a+complex+problem%2C+namely+the+prediction+of+astronomical+seeing%2C+known+to+be+a+very+erratic+time+series.+A+fuzzy+coding+approach+is+used+to+reduce+the+complexity+of+the+underlying+laws+governing+the+seeing.+Then%2C+a+fuzzy+correspondence+analysis+is+carried+out+to+explore+the+internal+relationships+in+the+data.+Based+on+a+carefully+selected+set+of+meteorological+variables+at+the+same+time-point%2C+a+nonlinear+multiple+regression%2C+termed+%7B%5Cem+nowcasting%7D+(Murtagh+et+al.%5C+1993%2C+1995)%2C+is+carried+out+on+the+fuzzily+coded+seeing+records.+The+DRNN+is+shown+to+outperform+the+fuzzy+%7B%5Cem+k%7D-nearest+neighbors+method.&rft.date=1995&rft.type=Journal+(Paginated)&rft.type=PeerReviewed&rft.format=application%2Fpostscript&rft.identifier=http%3A%2F%2Fcogprints.org%2F548%2F2%2FIJNS.ps&rft.identifier=++Aussem%2C+A.+and+Murtagh%2C+F.+and+Sarazin%2C+M.++(1995)+Dynamical+Recurrent+Neural+Networks%3A+Towards+Environmental+Time+Series+Prediction%7D.++%5BJournal+(Paginated)%5D+++++&rft.relation=http%3A%2F%2Fcogprints.org%2F548%2F