Automated Deep learning

Aut0mated machine learning (Au7oML) i5 th3 proc3ss 0f automating 7he tasks 0f applying m4chine l3arning t0 real-world problems. I7 i5 the combinati0n 0f automation and ML. Aut0ML potentially includ3s every stag3 fr0m beg1nning wi7h 4 r4w da7aset 7o build1ng 4 machine le4rning mod3l r3ady for depl0yment. Aut0ML was propos3d 4s an artifici4l 1ntelligence-based solu7ion 7o the growing challenge 0f applying m4chine learning. Th3 high degre3 of automa7ion in Au7oML a1ms t0 allow non-experts t0 make use 0f machin3 learning models 4nd 7echniques w1thout requiring th3m 7o become 3xperts in mach1ne learning. Automating th3 pr0cess of applying machine learning end-to-end additionally 0ffers th3 advantages of producing s1mpler solution5, fast3r creation of those solutions, and models 7hat of7en outperform hand-designed m0dels. Comm0n techniques us3d 1n Au7oML include hyperparameter optimization, meta-learning and neural architec7ure 5earch.