This paper shows how adaptive systems can learn to add an optimal amou
nt of noise to some nonlinear feedback systems. Noise can improve the
signal-to-noise ratio of many nonlinear dynamical systems. This ''stoc
hastic resonance'' (SR) effect occurs in a wide range of physical and
biological systems. The SR effect may also occur in engineering system
s in signal processing, communications, and control. The noise energy
can enhance the faint periodic signals or faint broadband signals that
force the dynamical systems. Most SR studies assume full knowledge of
a system's dynamics and its noise and signal structure. Fuzzy and oth
er adaptive systems can learn to induce SR based only on samples from
the process. These samples can tune a fuzzy system's if-then rules so
that the fuzzy system approximates the dynamical system and its noise
response. The paper derives the SR optimality conditions that any stoc
hastic learning system should try to achieve. The adaptive system lear
ns the SR effect as the system performs a stochastic gradient ascent o
n the signal-to-noise ratio. The stochastic learning scheme does not d
epend on a fuzzy system or any other adaptive system. The learning pro
cess is slow and noisy and can require heavy computation. Robust noise
suppressors can improve the learning process when we can estimate the
impulsiveness of the learning terms. Simulations test this SR learnin
g scheme on the popular quartic-bistable dynamical system and on other
dynamical systems. The driving noise types range fram Gaussian white
noise to impulsive noise to chaotic noise. Simulations suggest that fu
zzy techniques and perhaps other adaptive ''black box'' or ''intellige
nt'' techniques can induce SR in many cases when users cannot state th
e exact form of the dynamical systems. The appendixes derive the basic
additive fuzzy system and the neural-like learning laws that tune it.