Emerging　Technologies　are　often　viewed　as　breakthrough　innovations　that　radically　change　one　or　more　industries.　Many　point　out　that　these　＂new＂　technologies　are　evolved,　rather　than　created,　by　gradual　improvement　and　by　combining　existing　technologies　with　knowledge　and　techniques　in　multiple　fields.　This　paper　goes　further　by　interpreting　innovation　as　a　series　of　complex　and　dynamic　multi-agent　optimization　processes,　through　which　the　fittest　technologies　emerge　and　dominate.　The　＂optimization＂　concept　in　the　paper　emphasizes　improvement　(sometimes　feasibility)　rather　than　rigid　optimality,　and　centered　not　necessarily　upon　a　single,　mathematical　model.　This　broad　optimization　concept　views　innovation　as　complex　search,　learning,　design　and　selection　processes　which　can　all　be　optimized.　Technological　innovation　is　then　described　as　dynamic　multi-agent　optimization　processes　in　which　researchers,　educators,　inventors,　engineers,　managers,　producers,　investors,　traders,　customers　and　public　administrators　each　continuously　optimizes　one＇s　own　objectives　under　individual　resource　constraints,　while　cooperating　or　competing　with　other　actors.　Finally,　emerging　technologies　are　interpreted　as　prominent　outcomes　out　of　these　optimization　processes.　The　purpose　of　treating　innovation　as　special　optimization　processes　is　to　justify　and　recommend　the　use　of　more　recent　optimization　techniques　(e.　g.　genetic　algorithms　and　memetic　algorithms)　in　simulating　and　predicting　the　trends　in　emerging　technologies　development.