Reliable prediction intervals with directly optimized inductive conformal regression for deep learning.

By generating prediction intervals (PIs) to quantify the uncertainty of each prediction in deep learning regression, the risk of wrong predictions can be effectively controlled. High-quality PIs need to be as narrow as possible, whilst covering a preset proportion of real labels. At present, many approaches to improve the quality of PIs can effectively reduce the width of PIs, but they do not ensure that enough real labels are captured.

In field damage of high and low cyanogenic cassava due to a generalist insect herbivore Cyrtomenus bergi (Hemiptera: Cydnidae).

The hypothesis that cyanogenic potential in cassava roots deters polyphagous insects in the field is relevant to current efforts to reduce or eliminate the cyanogenic potential in cassava. To test this hypothesis, experiments were conducted in the field under natural selection pressure of the polyphagous root feeder Cyrtomenus bergi Froeschner (Hemiptera: Cydnidae). A number of cassava varieties (33) as well as 13 cassava siblings and their parental clone, each representing a determined level of cyanogenic potential (CNP), were scored for damage caused by C.

Cyanogenic potential in cassava and its influence on a generalist insect herbivore Cyrtomenus bergi (Hemiptera: Cydnidae).

The hypothesis that cyanogenic potential in cassava is a defense mechanism against arthropod pests is one of the crucial questions relevant to current efforts to reduce or eliminate cyanogenic potential (CNP) in cassava. The generalist arthropod Cyrtomenus bergi, which attacks cassava roots, was used in a bioassay relating oviposition and survival to CNP, concentration of nonglycosidic cyanogens, and linamarase (beta-glycosidase) activity in twelve selfed cassava siblings and their parental clone, which has segregated for different levels of cyanogenesis. Electron microscopic evaluation revealed an intracellular pathway of the stylet of C.