A machine’s ability to mimic human behavior is ARTIFICIAL INTELLIGENCE (AI). Machine Learning (ML) is a subset of AI, &Deep Learning (DL) is a subset of ML.ML provides systems the ability to automatically learn from experience without being explicitly programmed.Examples: Spam email detection, Online fraud detection, YouTube video recommendationsDL is ML which is capable of learning unsupervised from data that is unstructured or unlabeled.Examples: Self driving cars, Chatbots, DL RobotsThe primary difference between the two is the way we feed data to each. For example, to train an ML model on what a square is, we need to manually define the properties of a square like length, angle, etc. On the other hand, to train a DL model on the same, we just need to expose it to several images of different types of squares. DL models learn the features by themselves.In ML,even with the best feature specifications, it simply isn’t possible to grasp the complex patterns in the real world data. DL overcomes this limitation in ML. DL is very similar to how the human brain learns new concepts by being exposed to new data.
ML
DL
Hardware RequirementLow end machines will sufficeLarge storage, Advanced algorithms, High end computational power, High performance GPUsInput Data TypeStructured, LabeledRawSize of Input DataCan easily work with small amounts of dataDoesn't perform well with small amounts of dataModels’ Training TimeFew seconds to few hoursWeeksTesting TimeMoreLessDebuggingEasyTough/Almost impossibleDecision MakingTakes a decision and also gives us the reasoningResults can be similar to human work, but it won't tell us whyDecision PathYesNoAccuracyGoodBestExpenseLessMore- Jothi Abarna