In this course, you will delve into the core principles of AI, investigate its varied uses, and comprehend its role in revolutionizing various sectors.
 |
| Introduction to Artificial Intelligence (AI) | Coursera IBM |
Artificial intelligence (AI) is omnipresent, flawlessly woven into our everyday lives and professions. In this course, you will delve into the core principles of AI, investigate its varied uses, and comprehend its role in revolutionizing various sectors. You will get acquainted with the rudiments of generative AI and delve into its practical applications and use cases.
Notice!
Always refer to the module on your course for the most accurate and up-to-date information.
Attention!
If you have any questions that are not covered in this post, please feel free to leave them in the comments section below. Thank you for your engagement.
AI Concepts, Terminology, and Application Areas
1. Which of these statements is true?- Cognitive systems can derive mathematically precise answers following a rigid decision tree approach
- Cognitive systems can only process neatly organized structured data
- Cognitive systems can learn from their successes and failures
- Cognitive systems can only translate small volumes of audio data into their literal text translations at massive speeds
- Al is the subset of Data Science that uses Deep Learning algorithms on structured big data
- Artificial Intelligence and Machine Learning refer to the same thing since both the terms are often used interchangeably
-
Data Science is a subset of Al that uses machine learning algorithms to
extract meaning and dr
aw inferences from data - Deep Learning is a specialized subset of Machine Learning that uses layered neural networks to simulate human decision-making
- Takes data and rules as input and uses these inputs to develop an algorithm that will give us an answer
- Machine Learning algorithms are trained with large sets of datasets to determine the relationships between inputs and desired results to build the machine learning models.
- Machine Learning defines behavioral rules by comparing large data sets to find common patterns
- Machine Learning models can be continuously trained
- Takes data and answers as input and uses these inputs to create a set of rules that determine what the Machine Learning model will be
- It is useful for clustering data, where data is grouped according to how similar it is to its neighbors and dissimilar to everything else
- The algorithm ingests unlabeled data, draws inferences, and finds patterns from unstructured data
- It is useful for finding hidden patterns and or groupings in data and can be used to differentiate normal behavior with outliers such as fraudulent activity
- Takes data and rules as input and uses these inputs to develop an algorithm that will give us an answer
- Relies on providing the machine learning algorithm unlabeled data and letting the machine infer qualities
- Relies on providing the machine learning algorithm human-labeled data - the more samples you provide, the more precise the algorithm becomes in classifying new data
- Relies on providing the machine learning algorithm with a set of rules and constraints and letting it learn how to achieve its goals
- Tries its best to maximize its rewards by trying different combinations of allowed actions within the provided constraints
- Training subset is the data used to train the algorithm
- Validation data subset is used to validate results and fine-tune the algorithm's parameters
- Training data is used to fine-tune algorithm's parameters and evaluate how good the model is
- Testing data is data the model has never seen before and is used to evaluate how good the model is
- False
- True
- The process reduces errors over time by determining how far a given output is from the desired output
- The process works on the basis of reward and punishment depending on how close the given output is to the desired output
- An error function determines how far the given output is from the desired output
- Backpropagation uses training datasets to train neural networks to match known inputs to desired outputs
- Self Driving Vehicles
- Computer Vision
- Speech
- Natural Language
- Deconstructs sentences to decipher the context of use
- Ingests numerous samples of a person's voice until it can tell whether a new voice sample belongs to the same person
- Generates audio data and runs it through the network to see if it validates it as belonging to the subject
- Continues to correct the sample and run it through the classifier, repetitively, till an accurate voice sample is created
AI Issues, Ethics and Bias
1. Ethics in artificial intelligence is:- Something that somebody else is going to do in the future.
- Something that is entirely solved in current Al systems.
- Something that we need to apply today.
- Something that is not an issue.
2. One approach that helps developers avoid unintentionally creating bias in Al systems is:
- Using a wide variety of appropriately diverse data for training.
- Using highly specific training data from a narrow range.
- Not using any training data.
- None of the above.
- Data and insights belong to the people and businesses who created them. Organizations that collect, store, manage, or process data have an obligation to handle it responsibly.
- Knowing how an Al system arrives at an outcome is key to trust. To improve transparency, we should define how we build, deploy, and manage Al systems through scientific research.
- Unbiased models and a spirit of diversity and inclusion are necessary to create fair Al systems, which can mitigate, rather than magnify, our existing prejudices.
- Al can be applied to solve some of humanity's most pervasive problems and create opportunity for all.
- All of the above.
- Al systems in call centers providing context sensitive assistance to staff.
- Image recognition systems associating images of kitchens, shops, and laundry with women rather than men.
- Facial recognition systems performing well for individuals of all skin tones.
- Customers not being aware that they are interacting with a chatbot on a company website.
- Has very varied, unpredictable tasks.
- Features many repeatable tasks.
- Rules-based decision-making.
- Requires innovative problem solving.
- Not genuinely troubling, and the concern of very few Al experts.
- Something that can't be mitigated for.
- Short term and easily addressed when developing new Al systems.
- The concern of every Al developer, so they can be mitigated for as Al systems are developed.
- Concern about the trustworthiness of decision-making supported by Al systems.
- Loss of jobs due to Al replacing workers performing repetitive tasks.
- Privacy, for example, as human faces are photographed and recognised in public spaces.
- The necessity for any user of an Al system to have a high level of computer literacy.
- Racial, gender or other types of bias.
- In healthcare, Al is being used to interpret scans for early detection of cancer, eye disease, and other problems.
- Crime: to identify criminals before they commit a crime.
- In healthcare, Al is being used to predict where the next outbreak of a disease will occur.
- In agriculture, Al is being used to identify and recommend treatment for plant diseases.
- 165 million
- 7 million
- 133 million
- 48 million
- Performing regular tests and audits.
- Using only examples from their own environment as training data.
- Providing effective training data.
- Using government approved algorithms.
Final Assignment Part One
1. How would YOU define Al?
Your definition of Al can be similar or different from the ones given in the
course.
Al is a broad field that defines machine intelligence to perform certain
tasks, analyse data, identify patterns, extract relevant information etc.
2. Explain an application or use-case of Al that fascinates YOU.
It may or may not be something that is mentioned in the course.
One such case is using Al to analyzing time series data generated by lot
devices like example detect abnormalities in temperature or wind speed or
humidity
3. Pick a specific industry or an aspect of our lives or society and describe how YOU think it will be impacted by Artificial Intelligence in future.
What you discuss may or may not be something that is mentioned in the course.
For example in education entrance for first time applicants, Machine
Learning can be used to predict what sort of grades needed to enter a
course.
What is AI Applications and Examples of AI.
1. Which of the following is NOT a good way to define Al?- Al is Augmented Intelligence and is not intended to replace human intelligence rather extend human capabilities
- Al is the application of computing to solve problems in an intelligent way using algorithms.
- Al is all about machines replacing human intelligence.
- Al is the use of algorithms that enable computers to find patterns without humans having to hard code them manually
- Unsupervised Learning
- Reinforcement Learning
- Intuitive Learning
- Operate with human-level consciousness
- Perform independent tasks
- Learn new tasks to solve new problems
- Teach itself new strategies
- Philosophy
- All responses are correct
- Mathematics
- Statistics
- Self-Driving vehicles utilizing Computer Vision to navigate around objects
- Classifying rock samples to identify best places to drill for oil
- Collaborative Robots helping humans lift heavy containers
- Making precise patient diagnosis and prescribing independent treatment
- True
- False
- Real-time transcription
- On-demand online tutors
- Detecting cancerous moles in skin images
- Detecting fraudulent transactions
- Robots helping move items on shelves for stocking purposes
- Personal use in the home such as doing the laundry and cooking for example
- Robots helping humans lift heavy containers
- Robots assisting or replacing humans in jobs that may be dull, dangerous, ineffective or inefficient when done by humans
- An app on the mobile device that applies learnings from previous diagnosis data to assist the physicians in their current diagnoses
- Sending digital signals to a mobile device with a machine learning app via bluetooth
- Inserting a digitizer into the stethoscope tube to convert the analog sound of the heart beat into a digital signal
- Graphing heart beat data on the mobile device allowing a physician to spot trends
- Assisting patients with neurological damage by detecting patterns in massive movement related datasets and using robots to trigger specific movements in the human body to create new neural pathways in the brain.
- IBM Watson utilizing its information retrieval and natural language understanding capabilities to win the Jeopardy game show against its human competitors
- Watson analyzing Grammy nominated song lyrics over a 60 year period and categorizing them based on their emotion.
- Law enforcement authorities using facial recognition algorithms to identify suspects in multiple streams of video footage