MLT Unit 1 Part 2 Introduction

Que1.6. Write short note on well defined literacy problem with illustration. Answer Well defined literacy problem A computer program is said to learn from experience E with respect to some class of tasks T and performance measure P, if its performance at tasks in T, as measured by P, improves with experienceE. Three features in literacy problems 1. The class of tasks( T) 2. The measure of performance to be bettered( P) 3. The source of experience( E) For illustration 1. A checkers learning problem Task( T) Playing checkers. Performance measure( P) Percent of games won against opponents. Training experience( E) Playing practice games against itself. 2. A handwriting recognition literacy problem Task( T) Feting and classifying handwritten words within images. Performance measure( P) Percent of words rightly classified. Training experience( E) A database of handwritten words with given groups. 3. A robot driving literacy problem Task( T) Driving on public four- lane roadways using vision detectors. Performance measure( P) Average distance travelled before an error( as judged by mortal overseer). Training experience( E) A sequence of images and steering commands recorded while observing a mortal motorist. Que1.7. Describe well defined literacy problems part’s in machine literacy. Answer Well defined literacy problems part’s in machine literacy 1. Learning to fete spoken words Successful speech recognition systems employ machine literacy in some form. b. For illustration, the SPHINX system learns speaker-specific strategies for feting the primitive sounds( phonemes) and words from the observed speech signal. Neural network literacy styles and styles for learning retired Markov models are effective for automatically customizing to individual speakers, vocabularies, microphone characteristics, background noise,etc. 2. Learning to drive an independent vehicle Machine literacy styles have been used to train computer controlled vehicles to steer rightly when driving on a variety of road types. b. For illustration, the ALYINN system has used its learned strategies to drive unassisted at 70 long hauls per hour for 90 long hauls on public roadways among other buses . 3. Learning to classify new astronomical structures Machine literacy styles have been applied to a variety of large databases to learn general discrepancies implicit in the data. b. For illustration, decision tree learning algorithms have been used by NASA to learn how to classify elysian objects from the alternate Palomar Observatory Sky Survey. c. This system is used to automatically classify all objects in the Sky Survey, which consists of three terabytes of image data. 4. Learning to play world class backgammon a. The most successful computer programs for playing games similar as backgammon are grounded on machine literacy algorithms. b. For illustration, the world’s top computer program for backgammon, TD- GAMMON learned its strategy by playing over one million practice games against itself. Que1.8. Describe compactly the history of machine literacy. Answer Early history of machine literacy 1. In 1943, neurophysiologist Warren McCulloch and mathematician Walter Pitts wrote a paper about neurons, and how they work. They created a model of neurons using an electrical circuit, and therefore the neural network was created. 2. In 1952, Arthur Samuel created the first computer program which could learn as it ran. 3. Frank Rosenblatt designed the first artificial neural network in 1958, called Perceptron. The main thing of this was pattern and shape recognition. 4. In 1959, Bernard Widrow and Marcian Hoff created two models of neural network. The first was called ADELINE, and it could descry double patterns. For illustration, in a sluice of bits, it could prognosticate what the coming one would be. The second was called MADELINE, and it could exclude echo on phone lines. 1980s and 1990s 1. In 1982, John Hopfield suggested creating a network which had bidirectional lines, analogous to how neurons actually work. 2. Use of back propagation in neural networks came in 1986, when experimenters from the Stanford psychology department decided to extend an algorithm created by Widrow and Hoff in 1962. This allowed multiple layers to be used in a neural network, creating what are known as ‘ slow learners ’, which will learn over a long period of time. 3. In 1997, the IBM computer Deep Blue, which was a chess- playing computer, beat the world chess champion. 4. In 1998, exploration at AT&T Bell Laboratories on number recognition redounded in good delicacy in detecting handwritten postcodes from the US Postal Service. 21st Century 1. Since the launch of the 21st century, numerous businesses have realised that machine literacy will increase computation eventuality. This is why they are probing more heavily in it, in order to stay ahead of the competition. Machine Learning ways 1 – 11 L( CS/ IT- Sem- 5) 2. Some large systems include GoogleBrain( 2012) ii. AlexNet( 2012) iii. DeepFace( 2014) iv. DeepMind( 2014) OpenAI( 2015) vi. ResNet( 2015) vii. U-net( 2015) Que1.9. Explain compactly the term machine literacy. Answer 1. Machine literacy is an operation of Artificial Intelligence( AI) that provides systems the capability to automatically learn and ameliorate from experience without being explicitly programmed. 2. Machine literacy focuses on the development of computer programs that can pierce data. 3. The primary end is to allow the computers to learn automatically without mortal intervention or backing and acclimate conduct consequently. 4. Machine literacy enables analysis of massive amounts of data. 5. It generally delivers briskly and more accurate results in order to identify profitable openings or dangerous pitfalls. 6. Combining machine literacy with AI and cognitive technologies can make it indeed more effective in recycling large volumes of information. Que1.10. What are the operations of machine literacy? Answer Following are the operations of machine literacy 1. Image recognition Image recognition is the process of relating and detecting an object or a point in a digital image or videotape. b. This is used in numerous operations like systems for plant robotization, risk cell monitoring, and security surveillance. 2. Speech recognition Speech Recognition( SR) is the restatement of spoken words into textbook. It’s also known as Automatic Speech Recognition( ASR), computer speech recognition, or Speech To Text( STT). c. In speech recognition, a software operation recognizes spoken words. 3. Medical opinion ML provides styles, ways, and tools that can help in working individual and prognostic problems in a variety of medical disciplines. It’s being used for the analysis of the significance of clinical parameters and their combinations for prognostic. 4. Statistical arbitrage a. In finance, statistical arbitrage refers to automated trading strategies that are typical of a short- term and involve a large number of securities. b. In similar strategies, the stoner tries to apply a trading algorithm for a set of securities on the base of amounts similar as literal correlations and general profitable variables. 5. literacy associations Learning association is the process for discovering relations between variables in large data base. 6. birth Information birth( IE) is another operation of machine literacy. It’s the process of rooting structured information from unshaped data. Que1.11. What are the advantages and disadvantages of machine literacy? Answer Advantages of machine literacy are 1. fluently identifies trends and patterns Machine literacy can review large volumes of data and discover specific trends and patterns that would not be apparent to humans. b. For ane-commerce website like Flipkart, it serves to understand the browsing behaviours and purchase histories of its druggies to help cater to the right products, deals, and monuments applicable to them. c. It uses the results to reveal applicable announcements to them. 2. No mortal intervention demanded( robotization) Machine literacy does not bear physical force i.e., no mortal intervention is demanded. 3. nonstop enhancement ML algorithms gain experience, they keep perfecting in delicacy and effectiveness. b. As the quantum of data keeps growing, algorithms learn to make accurate prognostications briskly. 4. Handlingmulti-dimensional andmulti-variety data Machine literacy algorithms are good at handling data that are multi-dimensional andmulti-variety, and they can do this in dynamic or uncertain surroundings. Disadvantages of machine literacy are 1. Data accession Machine literacy requires massive data sets to train on, and these should be inclusive/ unprejudiced, and of good quality. 2. Time and coffers ML needs enough time to let the algorithms learn and develop enough to fulfill their purpose with a considerable quantum of delicacy and applicability. b. It also needs massive coffers to serve. 3. Interpretation of results To directly interpret results generated by the algorithms. We must precisely choose the algorithms for our purpose. 4. High error- vulnerability Machine literacy is independent but largely susceptible to crimes. b. It takes time to fete the source of the issue, and indeed longer to correct it.

Leave a Comment