What engineers need to know about Artificial Intelligence ?
Artificial intelligence (AI) systems by their natural are software - intensive. To create viable and trusted AI systems, engineers need technology and standards.
Following are the key aspects and elements of AI that engineers must understand to work with engineer system:
A. Introduction Concepts : AI, ML, and Deep Learning :
1. AI is defined as :
........the ability of Machines to perform tasks that normally required human intelligence - for example, recognizing patterns, learning for experience, drawing conclusions, making predictions, or taking actions - weather digitally or as the smart software behind autonomous physical systems. |
| Fig. of AI, ML & DL |
2. Machine learning (ML), a part of AI, is defined as :
A system that Learns and improve its performance at some task by using data and experience.3. Deep learning is defined as :
A family of machine learning techniques whole models extract important features by iteratively transforming the data "going deeper" toward meaningful pattern in the dataset with each transformation.Unlike traditional machine learning method, in which the creator of the model has to choose and encode features ahead of Time, Deep learning enabled a model to automatically learn features that matter.
B. AI Engineering Concept's :
i. AI depends on the human element :
1. AI arguments but does not replace human knowledge and expertise.
2. This basic understanding effects engineers of AI system in two-dimensional : human machine teaming and the probabilistic nature of AI "answer."
3. Engineer developing AI systems must account for the interaction between the system and the people who build and use it (human machine teaming).
4. Often, the success of those interactions comes down to trust and transparency.
5. Further, AI will produce Probabilistic answers.
6. How does the engineers of AI system know when a prediction is bad ?
ii. AI depends on labeled and unlabeled data as well as the system that store and access it :
1. The development in AI is due to the availability of data and the speed at which today's computers can process it.
2. AI system can classify, category and partition massive amounts of data to make the relevant information available for humans to analyze and make decisions.
3. Engineers must consider the data and the hardware and software system that support the data.
4. Large amounts of data require and computing environment that has the capacity to handle it.
5. Managing data required designing storage solutions around physical data constraints and type of queries desired.
iii. One AI, many algorithms :
1. When we take about AI, ML and Deep Learning we are referring to many different algorithms, many different approaches, not all of which are neural network based.
2. Many of the algorithms used in AI were generated in the 1950s, 1960s, & 1970s.
3. For example, the A* shortest path algorithm was conceived in the 1950s, and improved on the 1960s.
iv. The insight is the benefit of AI :
1. Engineer knows that it is impossible to test a system is every situation it will ever encounter.
2. An AI system can find an answer to never - seen - before situations that is insightful and have a very good Probability of being correct.
3. However, it is not necessarily correct, but probabilistic.
4. Thus, gaining increased confidence in AI is hard for engineers who need to focus on creating and validating a system.
v. An AI system depends on the system under which it runs :
1. When building a system that does not incorporate AI, you can build and test it in isolation.
2. Then deploy it and be certain it is going on behave just as it did in the lab.
3. An AI system depends on the conditions under which the AI runs and what the AI system is sensing, and this context adds another level of complexity.
0 Comments