A significant performance reduction is often observed in speech recognition when the rate of speech (ROS) is too low or too high. Most of present approaches to addressing the ROS variation focus on the change of speech signals in dynamic properties caused by ROS, and accordingly modify the dynamic model, e.g., the transition probabilities of the hidden Markov model (HMM). However, an abnormal R...

Supervised and unsupervised classification are common topics in machine learning in both scientific and industrial fields, which usually involve three tasks: prediction, exploration, and explanation. False discovery rate (FDR) theory has a close connection to classical classification theory, which must be employed in a sophisticated way to achieve good performance in various contexts. The study...

The learning rate is an information-theoretical quantity for bipartite Markov chains describing two coupled subsystems. It is defined as the rate at which transitions in the downstream subsystem tend to increase the mutual information between the two subsystems, and is bounded by the dissipation arising from these transitions. Its physical interpretation, however, is unclear, although it has be...

In gradient descent learning algorithms such as error backpropagation, the learning rate parameter can have a significant effect on generalization accuracy. In particular, decreasing the learning rate below that which yields the fastest convergence can significantly improve generalization accuracy, especially on large, complex problems. The learning rate also directly affects training speed, bu...

In the field of reinforcement learning, temporal difference (TD) learning is perhaps the most popular way to estimate the future discounted reward of states. We derive an equation for TD learning from statistical principles. Specifically, we start with the variational principle and then bootstrap to produce an updating rule for discounted state value estimates. The resulting equation is similar...

Two experiments assessed the contributions of implicit and explicit learning to base-rate sensitivity. Using a factorial design that included both implicit and explicit learning disruptions, we tested the hypothesis that implicit learning underlies base-rate sensitivity from experience (and that explicit learning contributes comparatively little). Participants learned to classify two categories...

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive o...

Zigler's hypothesis that mental age (MA) and not IQ determines the rate of learning is examined in the light of empirical evidence comparing the learning rates of normal and retarded children and young adults matched for MA. The results show that learning rate is a function of IQ as well as of MA. In general, children of average IQ learned serial and paired-associate lists significantly faster ...

The Learning Cost of Interest Rate Reversals* In this Paper, we suggest a new motivation for why central banks appear averse to reversing recent changes in their interest rate. We show, in a standard monetary model with forward-looking expectations, data uncertainty and parameter uncertainty, that there is a learning cost associated with interest rate reversals. A policy that frequently reverse...