This repository provides state of the art (SoTA) results for all machine learning problems. We do our best to keep this repository up to date. If you do find a problem's SoTA result is out of date or missing, please raise this as an issue or submit Google form (with this information: research paper name, dataset, metric, source code and year). We will fix it immediately.

Artificial Intelligence and Machine Learning have empowered our lives to a large extent. The number of advancements made in this space has revolutionized our society and continue making society a better place to live in. In terms of perception, both Artificial Intelligence and Machine Learning are often used in the same context which leads to confusion. AI is the concept in which machine makes smart decisions whereas Machine Learning is a sub-field of AI which makes decisions while learning patterns from the input data. In this blog, we would dissect each term and understand how Artificial Intelligence and Machine Learning are related to each other. What is Artificial Intelligence? The term Artificial Intelligence was recognized first in the year 1956 by John Mccarthy in an AI conference. In layman terms, Artificial Intelligence is about creating intelligent machines which could perform human-like actions. AI is not a modern-day phenomenon. In fact, it has been around since the advent of computers. The only thing that has changed is how we perceive AI and define its applications in the present world. The exponential growth of AI in the last decade or so has affected every sphere of our lives. Starting from a simple google search which gives the best results of a query to the creation of Siri or Alexa, one of the significant breakthroughs of the 21st century is Artificial Intelligence. The Four types of Artificial Intelligence are:- Reactive AI – This type of AI lacks historical data to perform actions, and completely reacts to a certain action taken at the moment. It works on the principle of Deep Reinforcement learning where a prize is awarded for any successful action and penalized vice versa. Google’s AlphaGo defeated experts in Go using this approach. Limited Memory – In the case of the limited memory, the past data is kept on adding to the memory. For example, in the case of selecting the best restaurant, the past locations would be taken into account and would be suggested accordingly. Theory of Mind – Such type of AI is yet to be built as it involves dealing with human emotions, and psychology. Face and gesture detection comes close but nothing advanced enough to understand human emotions. Self-Aware – This is the future advancement of AI which could configure self-representations. The machines could be conscious, and super-intelligent. Two of the most common usage of AI is in the field of Computer Vision, and Natural Language Processing. Computer Vision is the study of identifying objects such as Face Recognition, Real-time object detection, and so on. Detection of such movements could go a long way in analyzing the sentiments conveyed by a human being. Natural Language Processing, on the other hand, deals with textual data to extract insights or sentiments from it. From ChatBot Development to Speech Recognition like Amazon’s Alexa or Apple’s Siri all uses Natural Language to extract relevant meaning from the data. It is one of the widely popular fields of AI which has found its usefulness in every organization. One other application of AI which has gained popularity in recent times is the self-driving cars. It uses reinforcement learning technique to learn its best moves and identify the restrictions or blockage in front of the road. Many automobile companies are gradually adopting the concept of self-driving cars. What is Machine Learning? Machine Learning is a state-of-the-art subset of Artificial Intelligence which let machines learn from past data, and make accurate predictions. Machine Learning has been around for decades, and the first ML application that got popular was the Email Spam Filter Classification. The system is trained with a set of emails labeled as ‘spam’ and ‘not spam’ known as the training instance. Then a new set of unknown emails is fed to the trained system which then categorizes it as ‘spam’ or ‘not spam.’ All these predictions are made by a certain group of Regression, and Classification algorithms like – Linear Regression, Logistic Regression, Decision Tree, Random Forest, XGBoost, and so on. The usability of these algorithms varies based on the problem statement and the data set in operation. Along with these basic algorithms, a sub-field of Machine Learning which has gained immense popularity in recent times is Deep Learning. However, Deep Learning requires enormous computational power and works best with a massive amount of data. It uses neural networks whose architecture is similar to the human brain. Machine Learning could be subdivided into three categories – Supervised Learning – In supervised learning problems, both the input feature and the corresponding target variable is present in the dataset. Unsupervised Learning – The dataset is not labeled in an unsupervised learning problem i.e., only the input features are present, but not the target variable. The algorithms need to find out the separate clusters in the dataset based on certain patterns. Reinforcement Learning – In this type of problems, the learner is rewarded with a prize for every correct move, and penalized for every incorrect move. The application of Machine Learning is diversified in various domains like Banking, Healthcare, Retail, etc. One of the use cases in the banking industry is predicting the probability of credit loan default by a borrower given its past transactions, credit history, debt ratio, annual income, and so on. In Healthcare, Machine Learning is often been used to predict patient’s stay in the hospital, the likelihood of occurrence of a disease, identifying abnormal patterns in the cell, etc. Many software companies have incorporated Machine Learning in their workflow to steadfast the process of testing. Various manual, repetitive tasks are being replaced by machine learning models. Comparison Between AI and Machine Learning Machine Learning is the subset of Artificial Intelligence which has taken the advancement in AI to a whole new level. The thought behind letting the computer learn from themselves and voluminous data that are getting generated from various sources in the present world has led to the emergence of Machine Learning. In Machine Learning, the concept of neural networks plays a significant role in allowing the system to learn from themselves as well as maintaining its speed, and accuracy. The group of neural nets lets a model rectifying its prior decision and make a more accurate prediction next time. Artificial Intelligence is about acquiring knowledge and applying them to ensure success instead of accuracy. It makes the computer intelligent to make smart decisions on its own akin to the decisions made by a human being. The more complex the problem is, the better it is for AI to solve the complexity. On the other hand, Machine Learning is mostly about acquiring knowledge and maintaining better accuracy instead of success. The primary aim is to learn from the data to automate specific tasks. The possibilities around Machine Learning and Neural Networks are endless. A set of sentiments could be understood from raw text. A machine learning application could also listen to music, and even play a piece of appropriate music based on a person’s mood. NLP, a field of AI which has made some ground-breaking innovations in recent years uses Machine Learning to understand the nuances in natural language and learn to respond accordingly. Different sectors like banking, healthcare, manufacturing, etc., are reaping the benefits of Artificial Intelligence, particularly Machine Learning. Several tedious tasks are getting automated through ML which saves both time and money. Machine Learning has been sold these days consistently by marketers even before it has reached its full potential. AI could be seen as something of the old by the marketers who believe Machine Learning is the Holy Grail in the field of analytics. The future is not far when we would see human-like AI. The rapid advancement in technology has taken us closer than ever before to inevitability. The recent progress in the working AI is much down to how Machine Learning operates. Both Artificial Intelligence and Machine Learning has its own business applications and its usage is completely dependent on the requirements of an organization. AI is an age-old concept with Machine Learning picking up the pace in recent times. Companies like TCS, Infosys are yet to unleash the full potential of Machine Learning and trying to incorporate ML in their applications to keep pace with the rapidly growing Analytics space. Conclusion The hype around Artificial Intelligence and Machine Learning are such that various companies and even individuals want to master the skills without even knowing the difference between the two. Often both the terms are misused in the same context. To master Machine Learning, one needs to have a natural intuition about the data, ask the right questions, and find out the correct algorithms to use to build a model. It often doesn’t requiem how computational capacity. On the other hand, AI is about building intelligent systems which require advanced tools and techniques and often used in big companies like Google, Facebook, etc. There is a whole host of resources to master Machine Learning and AI. The Data Science blogs of Dimensionless is a good place to start with. Also, There are Online Data Science Courses which cover the various nitty gritty of Machine Learning.

Automatic Analysis of Music Performance Style One fundamental problem in computational music is analysis and modeling of performance style. Last year’s successful CUROP project revealed, through perceptual experiments, that players' control over rhythm is the strongest factor in the perceived quality of performance (already a publishable result). This year's project will hence investigate the computer analysis of the rhythmic component of performances in more detail, with the following aims: Implement and improve upon state-of-the-art beat detection methods. Carry out statistical analysis of rhythmic variation on a corpus of performances: Train a classifier into professional/amateur performance. Investigate to what extent rhythmic variations are controlled as opposed to random. Devise rhythmic style signatures of various performers for style recognition and retrieval. Investigate operations on rhythmic styles, e.g. apply Rachmaninoff's style to one's amateur recording. Solving the above problems is paramount to our understanding of what makes a good performance and what, quantitatively, are the differences between professional musician's styles. Applications include: musicology, teaching, automatic performance of music, high-level editing of music. This project requires integration of data mining, machine learning, and digital signal processing techniques, which are closely aligned with the expertise of the two supervisors: Dr Kirill Sidorov and Dr Andrew Jones. who are also experienced musicians. Via this project, the student will learn a variety of digital signal processing and machine learning techniques and will develop enhanced MATLAB programming skills, that are increasingly in demand for graduates. The student will work in our lab, with state-of-the-art facilities (powerful audio workstation, digital piano, audio gear). We will work collaboratively to ensure successful completion, including daily 30 minute meetings and longer weekly review meetings. The student will participate in the recently established Computational Music research sub-group. This project will contribute to longer-term development of this sub-group and foster new research avenues. Project Start/End Dates: Any 8 week period from 13th June 2016 to September 19th 2016. Contact/Supervisors: Kirill Sidorov Andrew Jones

Kernel machines such as the Support Vector Machine are widely used in solving machine learning problem, since they can approximate any function or decision boundary arbitrary well with enough training data. However, those methods applied on the kernel matrix (Gram matrix) of the data scale poorly with the size of the training dataset. The kernel trick may become intractable to compute as the computation and storage requirements for the kernel trick are exponentially proportional to the number of samples in the dataset. It takes a long time to train a model when training examples have big volume. For some specialized algorithms for linear Support Vector Machines, they operate much more quickly when the dimensionality of data is small because they operate on the covariance matrix rather than the kernel matrix of the training data. This paper we’ve chosen proposes a way to combine the advantages of the linear and nonlinear approaches. This method transformed the training and evaluation of any kernel machine by mapping the input data to a randomized low-dimensional feature space in order to create corresponding opera- tions of a linear machine. Those randomized features are designed to ensure that the inner products of the transformed data are nearly equal to those in the feature space of a user specific shift-invariant kernel. This method gives competitive results with state-of-the-art kernel-based classification and re- gression algorithms. What’s more, random features fix the problem of large scale of training data when computing the kernel matrix. The results have similar or even better testing error.

Rainfall is a form of precipitation and is responsible for providing most of the freshwater for animals and plants. Machine learning can be used to analyze data trends to develop a model. Deep learning on the other hand focuses more on using images specifically to analyze data. Trying to understand the patterns of rainfall to predict it has proven to be a difficult undertaking, as seen by the various research using machine learning and deep learning for this problem. When implementing a solution to this rainfall prediction problem, a vast amount of computational resources are usually required to execute it. Thus arises a need to properly store and analyze the data to effectively approach the prediction aspect. This paper investigated the comparison of predictive models for rainfall prediction using big data technologies and radar rainfall images. The literature on state-of-the-art prediction models was investigated and compared to survey which models could achieve satisfactory prediction results in combination with big data technologies. The models chosen were Random Forest Regressor and Deep LSTM and were used to predict 1,2, and 3 days ahead using monthly rainfall data. Results from this study showed that the Deep LSTM model performed better than the Random Forest Model for sequence lengths of 4, 8, and 12 when predicting 1, 2, and 3 months ahead.

The problem highlights the use of machine learning algorithms to categorize different comments scraped from an online platform and make relevant predictions about the topics associated with those comments. There are a total of 40 topics to classify these comments. Even though the problem seems like a simple classification problem, as we dive deeper to understand the data, we realize that the real problem asks us to make sense of the comments mentioned in the dataset and then assign categories. Since the number of topics/classes is much greater than any common classification problem, the expected accuracy won’t be too high. These days, Topic Modeling and Classification have received tremendous popularity when analyzing products and services for various brands, during election times to measure popularity, discover public sentiments around multiple issues, etc. Primarily deriving meaningful topics from these comments is incredibly challenging because of variations in language, insertion of emojis, and use of partial and profane comments. It is essential to choose a scheme that translates the comments to word embeddings to calculate some similarity between those comments to assign relevant topics; it is also imperative to translate the context and meaning of those comments and cluster them to relevant topics. There are multiple approaches to Topic Modeling, such as Latent Dirichlet Analysis (LDA) and Probabilistic Latent Semantic Analysis (LSA). These benchmark techniques utilized for such problems seem to provide viable results. The initial approach was to use Tf-Idf and Word2Vec to vectorize the comments and then use state-of-the-art classification techniques to assign topics to these vectors. When utilized, bag-of-Words with Tf-Idf and Word Embedding with Word2Vec would pose a significant hidden problem. The main problem with these approaches is that they treat the exact words with different meanings identically without adding any context to them. For example, the term “bank” in “Peter is fishing near the bank.” and “Two people robbed the state bank on Monday.” would have the same vectors in this representation. This approach would give us misleading results, and therefore, to improve the performance of our prediction mechanisms, it is essential to switch to a process that finds a way to translate the context of the words. Transformers: a reasonably new modeling technique, presented by Google’s research professionals in their seminal paper “Attention is All You Need,” tackles the exact problem. Google’s BERT (Bidirectional Encoder Representations from Transformers) combines ELMO context embedding and several Transformers, plus it’s bidirectional (which was a big novelty for Transformers). The vector assigned to a word using BERT is a function of the entire sentence; therefore, a word can have different vectors based on the context. ELMO is a word embedding technique that utilizes LSTMs to look at each sentence and then assigns those embeddings.

Recurrent neural networks (RNN) are the state-of-the-art algorithm for sequential data and are used by Apple's Siri and Google's voice search. It is an algorithm that remembers its input due to its internal memory, which makes the algorithm perfectly suited for solving machine learning problems involving sequential data. It is one of the algorithms that has great results in deep learning. In this article it is discussed how to predict the price of Bitcoin by analyzing the information of the last 6 years. We implemented a simple model that helps us better understand how time series works using Python and RNNs.

Video Summarization (VS) has been recognized as one of the most interested research and development field since the late 2000s. Generation of correct and adequate summaries for the given video is the end goal of the VS. There are different sub fields evolved since then such as Video Synopsis, Video Storytelling, Text-based Video Summaries (TVS), etc. Improvements in the Vision area with Convolutional Neural Network (CNN) approach have been accelerated this field further with all the ML categories such as Supervised Learning (SL), Unsupervised Learning (UL), and Reinforcement Learning (RL). Current State-of-the-Art (STOA) methods show that the usage of Natural Language Processing (NLP) and Transformer based solutions would make VS into a viable solution. However, the TVS area is yet to be investigated into the feasibility and real-world application. To fill this gap in TVS area, we introduce 3ML-TVS, called Three different Machine Learning to Text-based Video Summarization, a feasible solution that is made from the existing ML methods in Action Classification, Object Classification, and NLP. By fine-tuning each model individually, the result can be generated with promising accuracy. The proposed system is demonstrated the capacity of being applied to the real-world application also. This solution also proves that existing ML models have the capability to tackle much harder problem with simple systematic approaches rather than implementing a gigantic ML network.

No description available

This repository provides state of the art (SoTA) results for all machine learning problems. We do our best to keep this repository up to date. If you do find a problem's SoTA result is out of date or missing, please raise this as an issue or submit Google form (with this information: research paper name, dataset, metric, source code and year). We will fix it immediately.

This repository provides state of the art (SoTA) results for all machine learning problems. We do our best to keep th…

This repository provides state of the art (SoTA) results for all machine learning problems. We do our best to keep th…

This repository provides state of the art (SoTA) results for all machine learning problems. We do our best to keep th…

The course introduces a variety of central algorithms and methods essential for studies of statistical data analysis and machine learning. The course is project-based and through the various projects, the students are exposed to fundamental research problems in these fields, with the aim to reproduce state-of-the-art scientific results.

Multi-view modelling seeks to learn from multiple distinct feature sets and is a rapidly growing direction in machine learning. We foresee that this proJect will result in state-of-the-art performance for the clinical problem, as well as contribute towards meaningful research in the multi-view learning domain.

Reinforcement Learning (RL) is based on the idea where we learn by interacting with our environment. It is a collection of machine learning techniques where agent(s) learn how to behave in an environment by performing actions and assessing the results. RL approach is building programs that learn how to predict and act in stochastic environment, based on past experience. This course will provide an overview to students on some of the fundamental ideas on which modern RL is built including markov decision processes, value functions, monte carlo estimation, dynamic programming, TLD methods, approximation methods, Actor-Critic methods, etc. This course will help students to understand and apply RL in several systems including video distribution systems, game development, IOT devices, robotics, clinical decision making, industrial process control, finance portfolio balancing, etc. This subject aims to achieve the following goals:  To provide students with the knowledge to structure a reinforcement learning problem.  To introduce students to learn and apply basic RL algorithms for simple sequential decision-making problems in uncertain conditions  To introduce students research and development work in reinforcement learning by interpreting state-ofthe-art RL research and communicating their results.  To provide knowledge to students to build a RL system that knows how to make automated decisions.  To give students opportunities to understand the space of RL algorithms including Temporal Difference Learning, Monte Carlo, Q-Learning, approximation solution methods, A2C, A3C, etc.

# Image Classification using AWS SageMaker Use AWS Sagemaker to train a pretrained model that can perform image classification by using the Sagemaker profiling, debugger, hyperparameter tuning and other good ML engineering practices. This can be done on either the provided dog breed classication data set or one of your choice. ## Project Set Up and Installation Enter AWS through the gateway in the course and open SageMaker Studio. Download the starter files. Download/Make the dataset available. ## Dataset The provided dataset is the dogbreed classification dataset which can be found in the classroom. The project is designed to be dataset independent so if there is a dataset that is more interesting or relevant to your work, you are welcome to use it to complete the project. ### Access Upload the data to an S3 bucket through the AWS Gateway so that SageMaker has access to the data. ## Hyperparameter Tuning ResNet50 Model: * This project uses the ResNet-18 pre-trained model. * The ResNet-50 is a convolutional neural network that is 18 layers deep. This pre-trained version of the network is trained on more than a million images from the ImageNet database The pretrained network can classify images into 1000 object categories, such as keyboard, mouse, pencil, and many animals. * The dataset that we are going to use are an Image dataset which consist of images of dogs. * The dataset is divided into three parts training and validation and testing. For hyperparameters, I tuned the two following ones : - Learning rate : -> default is 0.001 and the chosen range is =```[0.0001, 0.1]``` -> learning rate is a ContinuousParameter. - epochs: -> defaut is 1e-08 and the chosen range is= ```[1e-9, 1e-8]``` -> the epochs is a IntegerParameter - Weight decay: -> default is 0.01 and the chosen range is = ```[1e-3, 1e-1]``` - Batch size : -> The chosen range is = ```[ 64, 128]``` -> batch-size is a CategoricalParameter **Best Training Jobs Hyperparameters after Tuning:**  **Optimizer: [Adam]** - Adam is a popular algorithm in the field of deep learning because it achieves good results fast. - learning_rate for Adam optimizer is a continuous parameter whose values are between ```[0.001, 0.01]``` - Adam is a popular algorithm in the field of deep learning because it achieves good results fast. - Computationally efficient. - Little memory requirements. - Also Well suited for problems that are large in terms of data and/or parameters. **Completed Training Jobs** **Logs Metrics During the Training Process** ## Debugging and Profiling - Debugging and Profiling was done with the help of the sagemaker.debugger module. - Amazon SageMaker Debugger provides full visibility into training jobs of state-of-the-art machine learning models. - This SageMaker Debugger module provides high-level methods to set up Debugger configurations to monitor, profile, and debug your training job. - Configure the Debugger-specific parameters when constructing a SageMaker estimator to gain visibility and insights into your training job.