As artificial intelligence becomes an essential part of our daily lives, the need for faster and more energy-efficient hardware continues to grow. In our latest research, conducted as part of the now-completed ORCA-LAB project, we explored how photonics\u2014using light instead of electricity\u2014can address these challenges.<\/p>\n\n\n\n
We focused on a key component of optical neural networks: \ud835\uddee\ud835\uddf0\ud835\ude01\ud835\uddf6\ud835\ude03\ud835\uddee\ud835\ude01\ud835\uddf6\ud835\uddfc\ud835\uddfb \ud835\uddf3\ud835\ude02\ud835\uddfb\ud835\uddf0\ud835\ude01\ud835\uddf6\ud835\uddfc\ud835\uddfb\ud835\ude00, which help the network learn and make decisions. This work demonstrates how \ud835\uddd9\ud835\uddee\ud835\uddef\ud835\uddff\ud835\ude06-\ud835\udde3\ud835\uddf2\ud835\uddff\ud835\uddfc\ud835\ude01 \ud835\uddf9\ud835\uddee\ud835\ude00\ud835\uddf2\ud835\uddff \ud835\uddf1\ud835\uddf6\ud835\uddfc\ud835\uddf1\ud835\uddf2\ud835\ude00 (FP-LDs) can serve as \ud835\uddf3\ud835\ude02\ud835\uddf9\ud835\uddf9\ud835\ude06 \ud835\uddfc\ud835\uddfd\ud835\ude01\ud835\uddf6\ud835\uddf0\ud835\uddee\ud835\uddf9, \ud835\uddfd\ud835\uddff\ud835\uddfc\ud835\uddf4\ud835\uddff\ud835\uddee\ud835\uddfa\ud835\uddfa\ud835\uddee\ud835\uddef\ud835\uddf9\ud835\uddf2 \ud835\uddee\ud835\uddf0\ud835\ude01\ud835\uddf6\ud835\ude03\ud835\uddee\ud835\ude01\ud835\uddf6\ud835\uddfc\ud835\uddfb \ud835\ude02\ud835\uddfb\ud835\uddf6\ud835\ude01\ud835\ude00\u2014a capability that has long been out of reach.<\/p>\n\n\n\n
Through simulations and experiments, we showed that these lasers respond to very short optical pulses (as short as 25 picoseconds) in complex and tunable ways. By adjusting experimental parameters\u2014such as injecting a secondary laser\u2014we were able to reproduce activation functions like \ud835\ude00\ud835\uddf6\ud835\uddf4\ud835\uddfa\ud835\uddfc\ud835\uddf6\ud835\uddf1 \ud835\uddee\ud835\uddfb\ud835\uddf1 \ud835\udde3\ud835\udde5\ud835\uddf2\ud835\udddf\ud835\udde8, entirely in the optical domain.<\/p>\n\n\n\n
Importantly, our solution works at high speeds (up to 10 GHz) and with extremely low energy consumption\u2014down to hundreds of femtojoules per operation\u2014making it a promising building block for future \ud835\ude02\ud835\uddf9\ud835\ude01\ud835\uddff\ud835\uddee\ud835\uddf3\ud835\uddee\ud835\ude00\ud835\ude01 \ud835\uddee\ud835\uddfb\ud835\uddf1 \ud835\uddf2\ud835\uddfb\ud835\uddf2\ud835\uddff\ud835\uddf4\ud835\ude06-\ud835\uddf2\ud835\uddf3\ud835\uddf3\ud835\uddf6\ud835\uddf0\ud835\uddf6\ud835\uddf2\ud835\uddfb\ud835\ude01 \ud835\uddfd\ud835\uddf5\ud835\uddfc\ud835\ude01\ud835\uddfc\ud835\uddfb\ud835\uddf6\ud835\uddf0 \ud835\uddfb\ud835\uddf2\ud835\ude02\ud835\uddff\ud835\uddee\ud835\uddf9 \ud835\uddfb\ud835\uddf2\ud835\ude01\ud835\ude04\ud835\uddfc\ud835\uddff\ud835\uddf8\ud835\ude00.<\/p>\n\n\n\n
Although the ORCA-LAB project has officially concluded, we are continuing to build on these results and explore their full potential in upcoming research activities.<\/p>\n\n\n\n
\ud83d\udcc4 Read the full paper here<\/p>\n\n\n\n
As artificial intelligence becomes an essential part of our daily lives, the need for faster and more energy-efficient hardware continues to grow. In our latest research, conducted as part of the now-completed ORCA-LAB project, we explored how photonics\u2014using light instead of electricity\u2014can address these challenges. We focused on a key component of optical neural networks:… Read More »\ud835\udde1\ud835\uddf2\ud835\ude04 \ud835\udde5\ud835\uddf2\ud835\ude00\ud835\ude02\ud835\uddf9\ud835\ude01\ud835\ude00 on Arxiv \ud835\uddf3\ud835\uddff\ud835\uddfc\ud835\uddfa \ud835\ude01\ud835\uddf5\ud835\uddf2 \ud835\udde2\ud835\udde5\ud835\uddd6\ud835\uddd4-\ud835\udddf\ud835\uddd4\ud835\uddd5 \ud835\udde3\ud835\uddff\ud835\uddfc\ud835\uddf7\ud835\uddf2\ud835\uddf0\ud835\ude01<\/span><\/a><\/p>\n","protected":false},"author":2,"featured_media":455,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"neve_meta_sidebar":"","neve_meta_container":"","neve_meta_enable_content_width":"","neve_meta_content_width":0,"neve_meta_title_alignment":"","neve_meta_author_avatar":"","neve_post_elements_order":"","neve_meta_disable_header":"","neve_meta_disable_footer":"","neve_meta_disable_title":""},"categories":[6],"tags":[],"_links":{"self":[{"href":"https:\/\/orca-lab.etf.bg.ac.rs\/index.php?rest_route=\/wp\/v2\/posts\/454"}],"collection":[{"href":"https:\/\/orca-lab.etf.bg.ac.rs\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/orca-lab.etf.bg.ac.rs\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/orca-lab.etf.bg.ac.rs\/index.php?rest_route=\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/orca-lab.etf.bg.ac.rs\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=454"}],"version-history":[{"count":2,"href":"https:\/\/orca-lab.etf.bg.ac.rs\/index.php?rest_route=\/wp\/v2\/posts\/454\/revisions"}],"predecessor-version":[{"id":457,"href":"https:\/\/orca-lab.etf.bg.ac.rs\/index.php?rest_route=\/wp\/v2\/posts\/454\/revisions\/457"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/orca-lab.etf.bg.ac.rs\/index.php?rest_route=\/wp\/v2\/media\/455"}],"wp:attachment":[{"href":"https:\/\/orca-lab.etf.bg.ac.rs\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=454"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/orca-lab.etf.bg.ac.rs\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=454"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/orca-lab.etf.bg.ac.rs\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=454"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}