![Reinforcement
learning
Reinforcement learning (RL) is an area of machine
learning concerned with how software agents ought to take
actions in an environment so as to maximize some notion of
cumulative reward. Reinforcement learning is one of three
basic machine learning paradigms, alongside supervised
learning and unsupervised learning.
It differs from supervised learning in that labelled
input/output pairs need not be presented, and sub-optimal
actions need not be explicitly corrected. Instead the focus is
finding a balance between exploration (of uncharted
territory) and exploitation (of current knowledge).[1]
The environment is typically formulated as a Markov
decision process (MDP), as many reinforcement learning
algorithms for this context utilize dynamic programming
techniques.](https://image.slidesharecdn.com/machinelearning-190930142440/85/Machine-learning-22-320.jpg)
Machine learning
- 1. Client Server Technology Presented by Mohit Modyani Bachelors of Computer Applications – Second Year Dezyne E’Cole College, Ajmer
- 2. Project report on Machine Learning Submitted To Dezyne E’Cole College Towards The Partial Fulfillment Of 2019 Year, Bachelor of Computer Application By Mohit Modyani Dezyne E’Cole College 106/10, Civil Line, Ajmer www.dezyneecole.com
- 3. Machine Learning Presented by: Mohit Modyani Bachelor of Compute Application- 2nd Year www.dezyneecole.com
- 4. Acknowledgement I am Mohit Modyani, Student of Bachelor Of Computer Application, Dezyne E’cole College. I would like to express my gratitude to each and every person who has contributed in encouraging me and helping me to coordinate my project. I also thank Dezyne E’cole College who provided insight and expertise that greatly assisted the project. A special thanks to my Teachers, Parents and Colleagues who have supported me at every step. Not to forget, the Almighty who blessed me with good health because of which I worked more efficiently and better.
- 5. CONTENTS • What Is Machine Learning ? • Methods of ML • Applications of ML • Challenges of ML • Future of ML • Supervised ML • Application of SML Speech Recognition Pattern Recognition Optical Character Recognition Handwriting Recognition Spam Recognition • Unsupervised Learning • Example of USML • Semi-Supervised Learning • Reinforcement Learning • Conclusion
- 6. WhAT is ML ? Machine learning is an application of artificial intelligence (AI) that provides systems the ability to automatically learn and improve from experience without being explicitly programmed.
- 7. Methods of ML Following is an overview of some of the most accepted ML methods – 1. Supervised Learning 2. Unsupervised Learning k-means clustering self-organizing maps value decomposition mapping of nearest neighbor 3. Semi-supervised Learning 4. Reinforcement Learning
- 8. Applications of ML 1. Financial Services 2. Marketing and Sales 3. Government 4. Expert System 5. Healthcare 6. Transportation 7. Oil and Gas The ML implementation in various sectors
- 9. Challegnes of ML 1. Memory networks 2. Natural language processing (NLP) 3. Understand deep nets training 4. One-shot learning 5. Deep reinforcement learning to control robots 6. Semantic segmentation 7. 111 Key challenges of Machine Learning
- 10. Future of ML 1. Fine-tuned personalization 2. Better search engine experiences 3. Evolution of data teams 4. No-code environments 5. Rise of quantum computing 5 ways it will impact everyday life
- 11. Supervised learning Supervised machine learning (SML) is the machine learning task of learning a function that maps an input to an output based on example input-output pairs. It infers a function from labelled training data consisting of a set of training examples. In supervised learning, each example is a pair consisting of an input object (typically a vector) and a desired output value (also called the supervisory signal). A supervised learning algorithm analyzes the training data and produces an inferred function, which can be used for mapping new examples. An optimal scenario will allow for the algorithm to correctly determine the class labels for unseen instances. This requires the learning algorithm to generalize from the training data to unseen situations in a "reasonable" way.
- 12. Example of smL Example: Classification: Machine is trained to classify something into some class. • classifying whether a patient has disease or not • classifying whether an email is spam or not Regression: Machine is trained to predict some value like price, weight or height. • predicting house/property price • predicting stock market price
- 13. Applications of smL The applications of supervised machine learning 1. Speech Recognition 2. Pattern Recognition 3. Optical Character Recognition 4. Spam Recognition
- 14. Speech Recognition The ability of a machine is to convert spoken language to text form, it is often used for voice dialing, call routing and voice search. EXAMPLE : Suppose this person has done some voice search. So, there are various software for speech recognition that convert voice into text form.
- 15. pattern Recognition It is the process of recognizing pattern by using machine learning algorithm. It classified data based on knowledge already gained or extracted from patterns. EXAMPLE : Nowadays latest smart phones have fingerprint scanners. So, we first save the pattern of our fingerprint and hen whenever we try to open our phone it recognize the pattern which we saved. If the pattern matches then the phone opens otherwise not.
- 16. Optical character Recognition It converts scanned images to readable text. It is the mechanical or electric conversion of images of typed, handwriting into machine encoded text. EXAMPLE : According to the image first the documents are scanned in the form of an image, then the documents are saved. Next, the OCR takes place and convert the scanned images into readable text.
- 17. handwriting Recognition A handwriting recognition system sensor can include a touch sensitive writing surface or a pen that contains sensor that detect angle, pressure and direction. The software then detect the writing pad and translates that writing in computer. EXAMPLE : When we go for license making there is a touch sensitive writing surface or a pen that contains sensor. We do our signature on that surface and it senses the angle, pressure and direction and translates that signature in computer.
- 18. sPam Recognition It is used to detect unwanted and useless emails and prevent those message from getting to users inbox. EXAMPLE : Suppose an email was send. First machine will check that message is useless or not, if email is useless or unwanted machine learning sent that email into spam but if email is wanted and not useless that it will send to the users inbox.
- 19. Unsupervised learning Unsupervised machine learning (USML) is a type of self-organized Hebbian learning that helps find previously unknown patterns in data set without pre-existing labels. It is also known as self-organization and allows modelling probability densities of given inputs. It is one of the main three categories of machine learning, along with supervised and reinforcement learning. Semi-supervised learning has also been described, and is a hybridization of supervised and unsupervised techniques.
- 20. Example of usmL Example: Clustering: A clustering problem is where you want to discover the inherent groupings is the data • such as grouping customers by purchasing behavior not Association: An association rule learning problem is where you want to discover rules that describe large portion of your data • such a people that buy X also tend to buy Y
- 21. Semi-supervised learning Semi-supervised machine learning (SSML) is a class of machine learning tasks and techniques that also make use of unlabelled data for training – typically a small amount of labelled data with a large amount of unlabelled data. Semi- supervised learning falls between unsupervised learning (without any labelled training data) and supervised learning (with completely labelled training data). Many machine- learning researchers have found that unlabelled data, when used in conjunction with a small amount of labelled data, can produce considerable improvement in learning accuracy. The cost associated with the labelling process thus may render a fully labelled training set infeasible, whereas acquisition of unlabelled data is relatively inexpensive. In such situations, semi-supervised learning can be of great practical value. Semi-supervised learning is also of theoretical interest in machine learning and as a model for human learning.
- 22. Reinforcement learning Reinforcement learning (RL) is an area of machine learning concerned with how software agents ought to take actions in an environment so as to maximize some notion of cumulative reward. Reinforcement learning is one of three basic machine learning paradigms, alongside supervised learning and unsupervised learning. It differs from supervised learning in that labelled input/output pairs need not be presented, and sub-optimal actions need not be explicitly corrected. Instead the focus is finding a balance between exploration (of uncharted territory) and exploitation (of current knowledge).[1] The environment is typically formulated as a Markov decision process (MDP), as many reinforcement learning algorithms for this context utilize dynamic programming techniques.
- 23. Conclusion of ML Machine Learning is a technique of training machines to perform the activities a human brain can do. But bit faster and better than an average human-being. Today we have seen that the machines can beat human champions in games such as Chess, AlphaGO, which are considered very complex. You have seen that machines can be trained to perform human activities in several areas and can aid humans in living better lives.
- 24. Thank you Mohit Modyani Bachelor of Computer Applications-Second Year Dezyne E’cole College www.dezyneecole.com