<h1 id="linear-time-invariance">Linear Time Invariance<a aria-hidden="true" class="anchor-heading icon-link" href="#linear-time-invariance"></a></h1>
<h1 id="lti-systems">LTI Systems<a aria-hidden="true" class="anchor-heading icon-link" href="#lti-systems"></a></h1>
<p>Linear Time Invariant systems are common in signal processing.</p>
<p>IDK if the systems we're modelling are linear or not, but time invariance is important.</p>
<p>In dynamic modelling, the conditions of the outputs at time point t are influenced by all the previous values of the input function, but never by the future values of the input function.</p>
<p>By default, operator networks don't have this constraint, meaning that changing the input function U(t) may actually change the predicted value for V(t-1), even though in the real world that makes no sense - you can't react to changes in the future!</p>
<h1 id="rq">RQ<a aria-hidden="true" class="anchor-heading icon-link" href="#rq"></a></h1>
<p>What changes need to be made to RBON architecture to constrain it to always model an LTI system?</p>
<p>How does this affect its accuracy in modelling LTI systems?</p>

Linear Time Invariance

thesis

# Notes

Browse topics using the links or search above.