This classification problem exemplifies the differences between the "hard" and "soft" learning strategies.
Figure 10 demonstrates the good agreement between the analytical solutions for the ensemble- averaged m u distribution, population average GFP response, and the ensemble average classifier performance and the simulation results for "soft" learning with γ = 1.
In the summer of 1990, Dennis Andrews, the head of Xerox's computer software division, X-Soft, learned of this work and asked product-development manager Jennifer Ware to visit PARC and look at the technology with an eye toward developing a product concept.
It's been described as having more laughs than most post-watershed comedies, and a superior form of "edutainment" — soft learning when school is out.
This analytical description is exact in the case of infinitely "soft" learning and infinite number of cells.
For instance, during the early stage of soft learning (i.e. active cluster merging), the algorithm could be applied to small 'chunks' of chromosomes.
We have developed an analytical theory that describes the training and the resulting performance of the cell-based classifier in the limit of "soft" learning (see Supporting Information Section 1).
Analytical treatment of soft learning; performance of the classifier model on the problem of discriminating toxicity data; classifier model based on experimentally derived PRM promoter response function; effect of non-uniformity of m u distribution in "master population".
