<html><body style="word-wrap: break-word; -webkit-nbsp-mode: space; line-break: after-white-space;"><head><meta http-equiv="Content-Type" content="text/html; charset=utf-8"></head><div class=""><blockquote type="cite" class=""><div class=""><div class="" style="word-wrap: break-word; -webkit-nbsp-mode: space; line-break: after-white-space;"><div class=""><div><blockquote type="cite" class=""><div class="" style="word-wrap: break-word; -webkit-nbsp-mode: space; line-break: after-white-space;"><div class=""><div class=""></div><div class=""><br class=""></div><div class=""></div><div class=""><br class=""></div><div class=""><br class=""></div><div class=""><b class="" style="font-size: 14px;"><font face="Times" class="">Wednesday, July 31, 2019 at 2:00pm</font></b></div><div class=""><b class="" style="font-size: 14px;"><font face="Times" class="">John Crerar Library, Room 390</font></b></div><div class=""><font face="Times" class=""><span class="" style="font-size: 14px;"><br class=""></span></font></div><div class=""><font face="Times" class=""><span class="" style="font-size: 14px;"><span class=""><b class="">Title:</b> </span><b class="">New Hardware Technology to Accelerate AI</b></span></font></div><div class=""><span class="" style="font-size: 14px;"><font face="Times" class=""><br class=""></font></span></div><div class=""><font face="Times" class=""><span class="" style="font-size: 14px;"><span class=""><b class="">Abstract:</b> </span><span class="" style="text-align: justify;">Deep neural networks are an excellent market opportunity for the computer industry.</span><span class="" style="text-align: justify;"> </span><span class="" style="text-align: justify;">Training them demands performance far beyond what CPUs can provide.</span><span class="" style="text-align: justify;"> </span><span class="" style="text-align: justify;">The demand has grown far faster than the growth in chip capability that Moore’s Law provides. GPUs do the job better than CPUs and for some time both training and inference have been a new and valuable market niche for the GPU.</span><span class="" style="text-align: justify;"> </span><span class="" style="text-align: justify;">But the GPU is not optimized for neural networks.</span><span class="" style="text-align: justify;"> </span><span class="" style="text-align: justify;">Now, new and better adapted architectures are appearing.</span><span class="" style="text-align: justify;"> </span><span class="" style="text-align: justify;">But the growth in demand for performance continues, and consequently the time to train large networks remains high, hours to days or even more.</span></span></font></div><p class="MsoNormal" style="text-align: justify;"><span class="" style="font-size: 14px;"><font face="Times" class=""><span class="">Cerebras Systems is a ventured-funded startup. We are designing a new system for training deep networks aimed at more performance than the fastest systems of today. I’ll indicate the new approaches in both architecture and hardware technology that allow us to do that<span class="" style="font-weight: bold;">.</span></span> </font></span></p><div class=""><font face="Times" class=""><span class="" style="font-size: 14px;"><b class="">Bio</b>: Rob Schreiber is a Distinguished Engineer at Cerebras Systems, Inc., where he works on architecture and programming of systems for accelerated training of deep neural networks. Schreiber’s research spans sequential and parallel algorithms for matrix computation, compiler optimization for parallel languages, and high performance computer design. With Moler and Gilbert, he developed the sparse matrix extension of Matlab. He created the NAS CG parallel benchmark. He was a designer of the High Performance Fortran language. Rob led the development at HP of the PICO system for synthesis of custom hardware accelerators. He has help pioneer the exploitation of photonic signaling in processors and networks. He is an ACM Fellow, a SIAM Fellow, and was awarded, in 2012, the Career Prize from the SIAM Activity Group in Supercomputing.</span></font></div><div class=""><font face="Times" class=""><span class="" style="font-size: 14px;"><br class=""></span></font></div><div class=""><span class="" style="font-size: 14px;"><font face="Times" class=""><b class="">Host:</b> Yanjing Li</font></span></div><div class=""><span class="" style="font-size: 14px;"><font face="Times" class=""><br class=""></font></span></div><div class=""><span class="" style="font-size: 14px;"><b class=""><font face="Times" class="">PDF:</font></b></span></div></div><div class=""></div></div></blockquote></div></div></div></div></blockquote></div></body></html>