Training recurrent neural networks: why and how? An illustration in dynamical process modeling.

The paper first summarizes a general approach to the training of recurrent neural networks by gradient-based algorithms, which leads to the introduction of four families of training algorithms. Because of the variety of possibilities thus available to the "neural network designer," the choice of the appropriate algorithm to solve a given problem becomes critical. We show that, in the case of process modeling, this choice depends on how noise interferes with the process to be modeled; this is evidenced by three examples of modeling of dynamical processes, where the detrimental effect of inappropriate training algorithms on the prediction error made by the network is clearly demonstrated.

[Our experience with the effectiveness and tolerance of ibuproxam in rheumatology]. / Nostra esperienza sull'efficacia e tollerabilità dell'ibuproxam in reumatologia.

The tolerance and effectiveness of Ibuproxam were investigated in 54 patients with arthrosis, R.A., ankylosing spondylitis, and fibrositis. Statistically significant reduction of pain at the end of the treatment was accompanied by decreased limitation of joint movement and morning stiffness. Simultaneous investigation of clinical and biological tolerance showed that the drug is infrequently responsible for side-effects that are always of negligible importance. Oesophagogastroduodenoscopy in subjects with a prior history and/or signs of active pathology of the upper digestive tract did not reveal exacerbation or recrudescence of lesions, nor the appearance of acute lesions.