<html><head><meta http-equiv="Content-Type" content="text/html charset=windows-1252"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;"><br><div><div>On Feb 10, 2014, at 4:51 PM, Brian J Mingus <<a href="mailto:brian.mingus@colorado.edu">brian.mingus@colorado.edu</a>> wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite"><div class="gmail_extra" style="font-family: Helvetica; font-size: 12px; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px;">That said, evolution is a blind designer. A human being can out-design billions of years of evolution in a few years with nice supercomputer and plenty of lab subjects. So, if your goal is to understand exactly what a human being is, you might study human development. But if your goal is to create something more sophisticated than a human without the annoyance of studying exactly how a human develops intelligence</div></blockquote>[a reasonable goal, depending on your research program]</div><div><br><blockquote type="cite"><div class="gmail_extra" style="font-family: Helvetica; font-size: 12px; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px;">, you might use deep networks with pretraining that automatically extract features that evolution baked in.</div><br class="Apple-interchange-newline"></blockquote></div><br><div>the key question is whether extracting features is, in itself, enough to replicate (or even better) the blind handiwork of evolution. my own guess is “absolutely not”. Evolution has a done a fine job with evo-crafting features — which deep networks might plausibly hope to match— but that there’s a lot of highly-selected circuitry <i>downstream</i> that probably cannot readily be captured through the mere acquisition of hierarchies of features. </div><div><br></div><div>On the left is a figure from <a href="http://www.frontiersin.org/Journal/10.3389/fnana.2011.00065/full">Solari and Stoner</a>’s magnificent cognitive consilience <a href="http://www.frontiersin.org/files/cognitiveconsilience/index.html">diagram</a>, which I encourage all students of cortical neuroscience to contemplate (click to zoom in). On the right is a figure representing Google’s cat detector, a state of the art unsupervised learner, yet still no match for humans when it comes to invariance or in the use of top-down visual information. Is the one on the right genuinely a useful approximation of the one of the left?</div><div> </div><div>In my own view there is an impedance mismatch between most current models and the intricacy of biological reality. </div><div><br></div><div>Cheers,</div><div>Gary </div><div><br></div><div><div style="margin: 0px;"><div style="margin: 0px;"><a href="http://www.frontiersin.org/files/cognitiveconsilience/index.html"><img apple-inline="yes" id="7587C1BF-E558-4293-B425-6AF5430D7F1F" height="308" width="436" apple-width="yes" apple-height="yes" src="cid:9B4EC889-CC32-4042-A3C6-12B1246F85C5@home"></a> <img apple-inline="yes" id="512FB149-1378-4DB0-89BB-2335B5A1F16F" height="300" width="359" apple-width="yes" apple-height="yes" src="cid:616507F4-F416-4141-BEF9-53EAE97AE870@home"></div><div style="margin: 0px;"><br></div><div style="margin: 0px;"> </div><div style="margin: 0px;"><br></div></div></div></body></html>