Neighborhood sequential and random training techniques for CMAC

An adaptive control algorithm based on Albus' CMAC (Cerebellar Model Articulation Controller) was studied with emphasis on how to train CMAC systems. Two training techniques-neighborhood sequential training and random training, have been devised. These techniques were used to generate mathematical functions, and both methods successfully circumvented the training interference resulting from CMAC's inherent generalization property. In the neighborhood sequential training method, a strategy was devised to utilize the discrete, finite state nature of the CMAC's address space for selecting points in the input space which would train CMAC systems in the most rapid manner possible. The random training method was found to converge on the training function with the greatest precision, although it requires longer training periods than the neighborhood sequential training method to achieve a desired performance level.     

Success rates with gamete intrafallopian transfer and in vitro fertilization in women of advanced maternal age

OBJECTIVE: To evaluate the effect of maternal age on outcomes for IVF and GIFT in women 40 to 45 years of age. DESIGN: Retrospective. SETTING: Boston IVF, a free-standing university-affiliated IVF and GIFT unit. PATIENTS: A total of 2,931 cycles of IVF and 1,826 cycles of GIFT were analyzed in women undergoing assisted reproductive technologies (IVF or GIFT) using autologous eggs. INTERVENTIONS: Medical records of patient outcomes were reviewed. RESULTS: For patients undergoing IVF, the cancellation rate for initiated cycles showed significant differences in women aged 25 to 39 (38.3%), women aged 40 to 43 (49.5%), and women aged 44 to 45 years (69.5%). A significantly lower delivery rate per stimulation and delivery rate per retrieval was found in women aged 40 to 43 years when compared with women aged 25 to 39 years. No deliveries occurred in 59 cycles in women aged 44 to 45 years, thereby representing a significant difference when compared with both women aged 25 to 39 years and women aged 40 to 43 years. For patients undergoing GIFT, the cancellation rate for initiated cycles was significantly higher in women aged 40 to 43 (25.0%) and 44 to 45 years (31.0%) when compared with women aged 25 to 39 years (15.1%). A significantly lower delivery rate per stimulation and delivery rate per retrieval was found in women aged 40 to 43 and 44 to 45 years when compared with women aged 25 to 39 years. CONCLUSIONS: Success rates for IVF and GIFT decline significantly in women > 40 years old. Women aged > or = 44 years are unlikely to benefit from the use of IVF and GIFT.     

A lysimeter system used to study the effect of transient waterlogging on crop growth and yield

The need for information on the effects of temporary waterlogging on crop growth to improve the basis of the design of field drainage systems is discussed, and the experimental facilities used for obtaining such information are described. These consist of 64 monolith lysimeters (32 of each of two soil types), 80 cm diameter and 135 cm deep, with facilities for control of the depth of the water-table and of rainfall. A movable glasshouse excludes unwanted rainfall and an irrigation system makes waterlogging of the soil possible when required. The crop environment and yields on the lysimeters are within the range expected under ordinary field conditions in England.