MACHINE LEARNING & DEEP LEARNINGYuan Xin (154405a) | Ryan (152338j) | Wayne (152107w)Table Of Contents?Executive Summary 4Definition and relationships 6What is Machine Learning 7What is Deep Learning 7What is Artificial Intelligence (AI) 7Relationships 8Essential steps of Machine Learning 9Gathering data 9Choosing a model 9Training 9Prediction 9Common methods of Machine Learning 10Supervised Learning 10Unsupervised Learning 10Neural Network 11What is a Neural Network 11Simple Neural Network and Deep Neural Network 11Consideration when choosing 12Machine Learning 12Predefined Features 12Deep Learning 12No Predefined Features 12Complex Features 12High Training Data 12High Computational Power 12Market Drivers 13Data storage become more affordable 13Rising trends of recommendation system 13Increased volume of diverse data 13Adoption of Internet Of Things 13Importance 14Why is it important 14Classroom Discussion 15What is inside the hidden layer 15Bayes Theorem in Machine Learning 15Get into Machine Learning without coding 15ML or DL is widely used in future 15Stakeholders 16Users of DL 16Users of ML 16Impacts on users 16Customer Service 16Customer Analytics 16Prevent Scams and Frauds 16Impacts on social politics 17ML VS Humans 17Pros of ML 17Cons of ML 17References 18?Executive SummaryIn order for computers to learn automatically without human intervention or assistance and adjust actions accordingly, it is undeniable that Machine Learning plays an important role in making that happen. Over the years, Machine Learning have been widely adopted and use for areas such as developing a recommendation system, assisting organization to make an important decision, as well as, detecting anomalies.Ryan ….
There are companies that invests in Machine Learning and Deep Learning to impact the lives of humans. Impacts can be on both individual and on social politics which may lead for the better or for the worse based on how it is used. Pros and Cons on using it can be balanced if it is used properly.Definition and relationshipsWhat is Machine Learning Machine learning is a field of computer science that gives computers the ability to learn without being explicitly programmed. In other words, computers can learn automatically without human intervention or assistance and adjust actions accordingly. Also, it allows computers to improve itself by exposing to new data over time.What is Deep Learning Deep learning, also known as deep structured learning / hierarchical learning, is part of a broader family of machine learning methods based on learning data representations, as opposed to task-specific algorithms.
Deep Learning is almost the same as Machine Learning, just that it is going “deeper”, which allows computer to learn more complex feature through multiple levels of abstraction using the Deep Neural NetworkWhat is Artificial Intelligence (AI)Artificial intelligence (AI, also machine intelligence, MI) is intelligence displayed by machines, in contrast with the natural intelligence (NI) displayed by humans and other animals.RelationshipsDL ? ML ? AIArtificial intelligence is any technique which enables computers to mimic human behaviour.Machine Learning is a subset of AI techniques which use statiscal methods to enable machines to improve with experience. Deep Learning is a subset of ML which make the computation of multi-layer neural networks feasible.Essential steps of Machine LearningGathering data Once you know exactly what you want and the equipments are in hand, it takes you to the first real step of machine learning- Gathering Data. This step is very crucial as the quality and quantity of data gathered will directly determine how good the predictive model will turn out to be. The data collected is then tabulated and called as Training Data.Choosing a model The next step that follows in the workflow is choosing a model among the many that researchers and data scientists have created over the years.
Make the choice of the right one that should get the job done.Training After the before steps are completed, you then move onto what is often considered the bulk of machine learning called training where the data is used to incrementally improve the model’s ability to predict. The training process involves initializing some random values for say A and B of our model, predict the output with those values, then compare it with the model’s prediction and then adjust the values so that they match the predictions that were made previously. This process then repeats and each cycle of updating is called one training step.PredictionMachine learning is basically using data to answer questions. So this is the final step where you get to answer few questions. This is the point where the value of machine learning is realized. Here you can Finally use your model to predict the outcome of what you want.
The above-mentioned steps take you from where you create a model to where you Predict its output and thus acts as a learning pathCommon methods of Machine LearningSupervised LearningThe system is given a set of labelled cases (training set) and asked to create a generalized model on those to act on unseen cases.Normally, the knowledge of output is known, with the presence of a teacher or expert. And the primary goal of Supervised learning is to predict class or value using algorithms such as Neural Network, Support Vector Machines, Decision Trees, Bayesian Classifiers and ect.Unsupervised Learning The system is given a set of unlabelled cases, and asked to find a pattern in them. Good for discovering hidden patterns.The knowledge of output is unknown and the primary goal is to determine data patterns or groupings.
Without the presence of a teacher or expert, self-guided algorithms such as K-means, Genetic Algorithms, Clustering Approaches and ect are adopted in Unsupervised Learning.Neural NetworkWhat is a Neural NetworkA neural network is essentially a series of mathematical and trainable units (algorithms) that is modelled after the “neurons” in the human brain. A “neuron” in a neural network is a simple mathematical function capturing and organizing information according to an architecture.
Neural networks have the ability to adapt to changing input so the network produces the best possible result without the need to redesign the output criteria.Simple Neural Network and Deep Neural NetworkTraditional Machine Learning uses Simple Neural Network. And Deep Learning uses Deep Neural Network.Simple Neural Network is a network that can use any network such as feedforward or recurrent network having 1 or 2 hidden layers.
But, when the number of hidden layers increases i.e. more than 2 than that is known as Deep Neural Network. Simple Neural Network is less complex and requires more information about features for performing feature selection and feature engineering method. On the other hand, Deep Neural Network does not require any information about features rather they perform optimum model tuning and model selection on their own.
Consideration when choosingMachine LearningPredefined FeaturesIf you know specifically which area you are focusing and are able to predefined your features i.e plants, foods, vehicles, alcohols and ect, you are fine to proceed with Machine Learning. Deep LearningNo Predefined FeaturesDeep learning algorithm would be useful to create new features if necessary and it is also very efficient in extracting out the perfect features.Complex FeaturesDeep learning is using layers of abstraction to help the computer learn more complex and complicated features.High Training Data To ensure the delivery of high performance and accuracy in Deep Learning, a minimum of 100k data is required. On the other hand, around 10k data would be sufficient for a Machine Learning System.High Computational Power Training a deep learning model can take a long time, from days to weeks.
Using GPU acceleration can speed up the process significantly. High-performance GPUs have a parallel architecture that is efficient for Deep Learning.To summarise, Deep Learning algorithms heavily depend on high-end machines (GPU). On the other hand, Traditional Machine Learning algorithms can work on low-end machines (CPU).Market DriversData storage become more affordableWith Cloud Computing, Cloud services for data back-up and storage are increasingly common and becoming more affordable. Huge amount of open source data are available on the cloud, driving developers to share these resources and use these data for training their models.Rising trends of recommendation systemRecommendation systems are used across a wide range of industries, most notably online shopping sites which help organisation to get deeper insights about the customer behavior by helping them discover new and relevant offers thereby leading to stronger customer relationships and higher sales for the business.
Increased volume of diverse dataWith the increase volume of data (Big Data) to be handled quantitatively, organization are using Machine Learning to extract out features and identify patterns and trends, so they could make better decisions.Adoption of Internet Of ThingsIoT devices are generating tons of data, and machine learning is being employed to analyze and peruse that data to help improve efficiency and customer service, and reduce costs and energy consumption.ImportanceWhy is it importantMachine learning has several very practical applications that drive the kind of real business results, such as time and money savings. It has the potential to dramatically impact the future of your organization.
Also, it is going to have a huge effect on politics, the economy, and society as a whole. Entire industries will be automated, leaving millions of people out of work and unable to retrain fast enough to stay ahead of ever-faster technological improvements.Classroom DiscussionWhat is inside the hidden layerThe inputs feed into a layer of hidden units, which can feed into layers of more hidden units, which eventually feed into the output layer. Each of the hidden units is a squashed linear function of its inputs. Neural networks of this type can have as inputs any real numbers, and they have a real number as output.Bayes Theorem in Machine LearningBayes’ Theorem is the fundamental result of probability theory – it puts the posterior probability P(H|D) of a hypothesis as a product of the probability of the data given the hypothesis(P(D|H)), multiplied by the probability of the hypothesis (P(H)), divided by the probability of seeing the data. (P(D)) We have already seen one application of Bayes Theorem in class – in the analysis of Information Cascades, we have found that it is possible for rational decisions to be made where one’s own personal information is discarded, based upon the conditional probabilities calculated via Bayes’ Theorem. Such a powerful concept, and succinctly summarized in the following formula, , Bayes’ Theorem is the basis of a branch of Machine Learning – that is, of the Bayesian variety.
Get into Machine Learning without codingCustom Vision Service is a tool for easily training, deploying, and improving custom image classifiers. With just a handful of images per category, you can train your own image classifier in minutes (No coding required).ML or DL is widely used in futureDue to high computational power and data required in Deep Learning compared to Machine Learning, ML will still be more widely used as it do not heavily depend on high-end machine. Using affordable machines and data, most of us will still be able to implement Machine Learning in our applications or projects.StakeholdersUsers of DLBaylabs, Graphcore, Cerebras, Visenze, Deep VisionUsers of MLGoogle, Pinterest, Soundhound, Zebra, IBM, BaiduImpacts on usersCustomer ServiceBy giving the wrong results to a user based on his input, reviews on the app will be all negative reviews.
However with ML, customers can now speak in their own words to the app and the app will be able to understand more quickly and know what the customer needs. This solves their problem faster resulting in better customer service.Customer AnalyticsThey say insanity is doing the same thing over and over and over again. Well analytics are here to make us all a little more sane. Analytics takes information from customer data and uses it to predict future trends and behavior patterns. From a customer service perspective predictive analytics help anticipate when a customer that’s shopping on your website will need agent help.