So, deep learning is considered to be a part of machine learning. This diagram very nicely depicts what deep learning is at a very high level. You have the all-encompassing artificial intelligence, which is more a concept rather than a technology or a technical concept, right? So, it is more of a concept at a very high level. Artificial intelligence, under the hood, is actually machine learning and deep learning. Machine learning is a broader concept, you can say, or a broader technology, and deep learning is a subset of machine learning. The primary difference between machine learning and deep learning is that deep learning uses neural networks, and it is suitable for handling large amounts of unstructured data. The last but not least, one of the major differences between machine learning and deep learning is that in machine learning, the feature extraction or the feature engineering is done by the data scientists manually, but in deep learning, since we use neural networks, the feature engineering happens automatically. So, that's a little bit of a quick difference between machine learning and deep learning, and this diagram very nicely depicts the relation between artificial intelligence, machine learning, and deep learning. Now, why do we need deep learning? Machine learning was there for quite some time, and it can do a lot of stuff that probably deep learning can do. But it's not very good at handling large amounts of unstructured data like images, voice, or even text, for that matter. So, traditional machine learning is not that good at doing this. Traditional machine learning can handle large amounts of structured data, but when it comes to unstructured data, it's a big challenge. So, that is one of the key differentiators for deep learning. That is number one. And increasingly, for artificial intelligence, we need image recognition, and we need to process and analyze images and voice. That's the reason deep learning is required compared to, let's say, traditional machine learning. It can also perform complex algorithms, more complex than, let's say, what machine learning can do, and it can achieve the best performance with large amounts of data. So, the more data you have, let's say reference data or labeled data, the better the system will do because the training process will be that much better. And last but not least, with deep learning, you can really avoid the manual process of feature extraction. Those are some of the reasons why we need deep learning. Some of the applications of deep learning: Deep learning has made major inroads, and a major area in which deep learning is applied is healthcare. And within healthcare, particularly oncology, which is basically cancer-related stuff. One of the issues with cancer is that a lot of cancers today are curable; they can be cured if they are detected early on. And the challenge with that is when a diagnostic is performed, let's say an image has been taken of a patient to detect whether there is cancer or not, you need a specialist to look at the image and determine whether the patient is fine or if there is any onset of cancer. The number of specialists is limited. So, if we use deep learning, if we use automation here, or if we use artificial intelligence here, then the system can, with a certain amount of good accuracy, determine whether a particular patient has cancer or not. So, the prediction or the detection process of a disease like cancer can be expedited. The detection process can be expedited and can be faster, without really waiting for a specialist. We can obviously, then, once the artificial intelligence detects or predicts that there is an onset of cancer, this can be cross-checked by a doctor. But at least the initial screening process can be automated, and that is where the current focus is with respect to deep learning in healthcare. what else robotics is another area deep learning is majorly used in robotics and you must have seen nowadays robots are everywhere humanoids the industrial robots which are used for manufacturing process you must have heard about sofia who got citizenship with saudi arabia and so on there are multiple such robots which are knowledge oriented but there are also industrial robots are used in industries in the manufacturing process and increasingly in security and also in defense for example image processing video is fed to them and they need to be able to detect objects obstacles and so on and so forth so that's where deep learning is used they need to be able to hear and make sense of the sounds that they are hearing that needs deep learning as well so robotics is a major area where deep learning is applied Then we have self-driving cars or autonomous cars. You must have heard of Google's autonomous car, which has been tested for millions of miles and is pretty much incident-free. There were, of course, a couple of incidents here and there, but it is considered to be fairly safe. And there are today a lot of automotive companies; in fact, pretty much every automotive company worth its name is investing in self-driving cars or autonomous cars. It is predicted that in the next probably 10 to 15 years, these will be in production and will be used extensively in real life. Right now, they are all in R&D and in test phases, but pretty soon, these will be on the road. So, this is another area where deep learning is used. And how is it used? Where is it used? Within autonomous driving, the car actually is fed with video of its surroundings, and it is supposed to process that information, process that video, and determine if there are any obstacles. It has to determine if there are any cars in the vicinity, detect whether it is driving in the lane, and also it has to determine whether the signal is green or red so that accordingly, it can move forward or wait. So, for all these video analyses, deep learning is used. In addition to that, the overall training to drive the car happens in a deep learning environment. So, again, a lot of scope here to use deep learning. A couple of other applications are machine translations. Today, we have a lot of information, and very often, this information is in one particular language, and more specifically in English. People need information in various parts of the world. It is pretty difficult for human beings to translate each and every piece of information or every document into all possible languages. There are probably at least hundreds of languages, or if not more. To translate each and every document into every language is pretty difficult. Therefore, we can use deep learning to do pretty much like a real-time translation mechanism. So, we don't have to translate everything and keep it ready, but we train applications or artificial intelligence systems that will do the translation on the fly. For example, you go somewhere like China and you want to know what is written on a signboard. Now, it is impossible for somebody to translate that and put it on the web or something like that. So, you have an application which is trained to translate stuff on the fly. This can probably be running on your mobile phone, on your smartphone. You scan this, and the application will instantly translate that from Chinese to English. That is one. Then, there could be web applications where there may be a research document which is all in maybe Chinese or Japanese, and you want to translate that to study that document or in that case, you need to translate. So, therefore, deep learning is used in such situations as well. And that is again on demand, so it is not like you have to translate all these documents from other languages into English in one shot and keep it somewhere. That is again pretty much an impossible task, but on a need basis, you have systems that are trained to translate on the fly. So, machine translation is another major area where deep learning is used.