Memory-based learning

It should be pointed out that this approach can t strictly be used for TTS purposes as acoustic features (e.g. time in seconds) measured from the corpus waveforms were used in addition to features that would be available at run time. Following this initial work, a number of studies have used decision trees [264] [418], and a wide variety of other machine learning algorithms have been applied to the problem including memory based learning [77] [402], Bayesian classifiers [516], support vector machines [87] and neural networks [157]. Similar results are reported in most cases, and it seems that the most important factors in the success of a system are the features used and the quality and quantity of data rather than the particular machine learning algorithm used. 

Busser, B., Daelemans, W., and van den Bosch, A. Predicting phrase breaks with memory-based learning. In 4th ISCA Tutorial and Research Workshop on Speech Synthesis (2001). 

Fosters active learning environment instead of memory-based learning... 

Lesioned animal models are based upon the assumption that the destruction of basal forebrain cholinergic neurons by injection of a neurotoxin, such as ibotenic acid, is sufficient to reproduce some of the cognitive impairments associated with AD, mainly memory and learning deficits. Indeed, pathology in the basal forebrain... 

Simple experimental procedure that leads to memorization by repetition or rote learning as the evaluation was mainly memory-based... 

The tests generally involve some form of maze but the simplest is the passive avoidance test. In this the animal learns that in a certain environment it will be punished with an electric shock for some particular action, like stepping onto a special part of the floor of the test chamber. The test of memory is how long the rat avoids (remains passive to) making the movement that will initiate the shock. Of course, drugs that reduce the animal s anxiety also modify the response. Using a maze in its simplest T shape, the animal is placed at the base of the vertical arm and a food reward at the end of one of the horizontal arms. Clearly the animal has to learn which arm contains the reward. Memory is assessed by the time taken for a food-deprived animal to reach the reward and the number of false arm entries. This simple system can be made more complex by introducing many more arms and branches but the principle is the same. 

By definition, learning is the acquisition of new information, whereas memory is the retention of acquired information. Driven by knowledge obtained by scientists before them, various disciples of neuroscience over the course of the past 100 years have learned more about various components of learning and memory, whether at the molecular level of synapses or at the systems level of brain circuitry. The concept of memory of mind has existed since the time of Aristotle. It is only during the past 50 years or so that scientists have begun to unravel some of the anatomical and cellular bases underlying such a complex mental process. 

---