Cloud Hosting

Machine learning has bestowed humanity the power to run tasks in an automated manner. It allows improving things that we already do by studying a continuous stream of data related to that same task. Machine learning has a wide array of applications that belongs to different fields, ranging from space research to digital marketing.

Machine learning also forms the basis of artificial intelligence. Were not yet flooded with machines capable of throwing judgments on their own. Its still a long way to reach there. But the possibilities generated along the way are endless.

So, it is the best time to pick up and learn machine learning. Of course, machine learning is a complex field but that doesnt mean that it cant be learned in an easy way. To help you through, here we are with our pick of the 20 best machine learning books:

Author Andriy BurkovLatest Edition FirstPublisher Andriy BurkovFormat ebook (Leanpub)/Hardcover/Paperback

Is it possible to explain various machine learning topics in a mere 100 pages? The Hundred-Page Machine Learning Book by Andriy Burkov is an effort to realize the same. Written in an easy-to-comprehend manner, the machine learning book is endorsed by reputed thought leaders to the likes of the Director of Research at Google, Peter Norvig and Sujeet Varakhedi, Head of Engineering at eBay. It is the best books for Machine Learning to start with.

Post a thorough reading of the book, you will be able to build and appreciate complex AI systems, clear an ML-based interview, and even start your very own ml-based business. The book, however, is not meant for absolute machine learning beginners. If youre looking for something more fundamental look somewhere else.

Topics covered

You can buy this book here.

Author Toby SegaranLatest Edition FirstPublisher OReilly MediaFormat Kindle/Paperback

Regarded among the best books to begin understanding machine learning, the Programming Collective Intelligence by Toby Segaran was written way before, in 2007, data science and machine learning reached its present status of top career avenues. The book makes use of Python as the vehicle of delivering the knowledge to its readers.

The Programming Collective Intelligence is less of an introduction to machine learning and more of a guide for implementing ml. The book details on creating efficient ml algorithms for gathering data from applications, creating programs for accessing data from websites, and inferring the gathered data. Each chapter features exercises for extending the stated algorithms and further improve their efficiency and effectiveness.

Topics covered

You can buy this book here.

Author Drew Conway and John Myles WhiteLatest Edition FirstPublisher OReilly MediaFormat Kindle/Paperback

The Machine Learning for Hackers book is meant for the experienced programmer interested in crunching data. Here, the word hackers refer to adroit mathematicians. As most of the book is based on data analysis in R, it is an excellent option for those with a good knowledge of R. The book also details using advanced R in data wrangling.

Perhaps the most important highlight of the Machine Learning for Hackers book is the inclusion of apposite case studies highlighting the importance of using machine learning algorithms. Rather than delving deeper into the mathematical theory of machine learning, the book explains numerous real-life examples to make learning ml easier and faster.

Topics covered

You can buy this book here.

Author Tom M. MitchellLatest Edition FirstPublisher McGraw Hill EducationFormat Paperback

Machine Learning by Tom M. Mitchell is a fitting book for getting started with machine learning. It offers a comprehensive overview of machine learning theorems with pseudocode summaries of the respective algorithms. The Machine Learning book is full of examples and case studies to ease a readers effort for learning and grasping ml algorithms.

If you wish to start your career in machine learning, then this book is a must-have. Thanks to a well-explained narrative, a thorough explanation of ml basics, and project-oriented homework assignments, the book on machine learning is a suitable candidate to be included in any machine learning course or program.

Topics covered

You can buy this book here.

Author Trevor Hastie, Robert Tibshirani, and Jerome FriedmanLatest Edition SecondPublisher SpringerFormat Hardcover/Kindle

If you like statistics and want to learn machine learning from the perspective of stats then The Elements of Statistical Learning is the book that you must read. The machine learning book emphasizes mathematical derivations for defining the underlying logic of an ml algorithm. Before picking up this book, ensure that you have at least a basic understanding of linear algebra.

The concepts explained in The Elements of Statistical Learning book arent beginner-friendly. Hence, you might find it complex to digest. If you still, however, want to learn them then you can check out the An Introduction to Statistical Learning book. It explains the same concepts but in a beginner-friendly way.

Topics covered

You can buy this book here.

Author Yaser Abu Mostafa, Malik Magdon-Ismail, and Hsuan-Tien LinLatest Edition FirstPublisher AMLBookFormat Hardcover/Kindle

Want to get a comprehensive introduction to machine learning in less time? And have a good understanding of engineering mathematics? Try the Learning from Data: A Short Coursebook. Instead of imparting knowledge about the various advanced concepts pertaining to machine learning, the book prepares its readers to better comprehend the complex machine learning concepts.

The Learning from Data: A Short Coursebook ditches lengthy and beating around the bush explanations for succinct, to the points explanations. To reinforce learning from this machine learning book, you can also refer to the online tutorials from the author Yaser Abu Mostafa.

Topics covered

You can buy this book here.

Author Christopher M. BishopLatest Edition SecondPublisher SpringerFormat Hardcover/Kindle/Paperback

Written by Christopher M. Bishop, the Pattern Recognition and Machine Learning book serves as an excellent reference for understanding and using statistical techniques in machine learning and pattern recognition. A sound understanding of linear algebra and multivariate calculus are prerequisites for going through the machine learning book.

The Pattern Recognition and Machine Learning book present detailed practice exercises for offering a comprehensive introduction to statistical pattern recognition techniques. The book leverages graphical models in a unique way of describing probability distributions. Though not mandatory, some experience with probability will hasten the learning process.

Topics covered

You can buy this book here.

Author Steven Bird, Ewan Klein, and Edward LoperLatest Edition FirstPublisher OReilly MediaFormat Available

Natural language processing is the backbone of machine learning systems. The Natural Language Processing with Python book uses the Python programming language to guide you into using NLTK, the popular suite of Python libraries and programs for symbolic and statistical natural language processing for English and NLP in general.

The Natural Language Processing with Python book presents powerful Python codes demonstrating NLP in a clear, precise manner. Readers are able to access well-annotated datasets for analyzing and dealing with unstructured data, linguistic structure in text, and other NLP-oriented aspects.

Topics covered

You can buy this book here.

Author David BarberLatest Edition FirstPublisher Cambridge University PressFormat Hardcover/Kindle/Paperback

For anyone interested in entering the field of machine learning, Bayesian Reasoning and Machine Learning is a must-have. The book is a fitting solution for computer scientists interested in learning ml but doesnt have a background in calculus and linear algebra.

There is no scarcity of well-explained examples and exercises in the Bayesian Reasoning and Machine Learning book. This makes the book also ideal for undergraduate and graduate computer science students. The machine learning book comes with additional online resources and a comprehensive software package that includes demos and teaching materials for instructors.

Topics covered

You can buy this book here.

Author Shai Shalev-Shwartz and Shai Ben-DavidLatest Edition FirstPublisher Cambridge University PressFormat Hardcover/Kindle/Paperback

The Understanding Machine Learning book offers a structured introduction to machine learning. The book dives into the fundamental theories and algorithmic paradigms of machine learning, and mathematical derivations.

The machine learning presents a wide array of machine learning topics in an easy-to-understand way. The Understanding Machine Learning book is fitting for anyone ranging from computer science students to non-expert readers in computer science, engineering, mathematics, and statistics.

Topics covered

You can buy this book here.

Author Oliver TheobaldLatest Edition SecondPublisher Scatterplot PressFormat Kindle/Paperback

Have no prior experience and exposure to machine learning? But still, want to learn it? Then you must not miss out on the Machine Learning for Absolute Beginners book by Oliver Theobald. Obviously, no coding or mathematical background is required to benefit from this machine learning book.

For anyone looking to get the most toned-down definition of machine learning and related concepts, the Machine Learning for Absolute Beginners book is one of the most fitting options. In order to ensure that the readers follow everything mentioned in the book easily, clear explanations and visual examples accompany various ml algorithms.

Topics covered

You can buy this book here.

Author John Paul Mueller and Luca MassaronLatest Edition FirstPublisher For DummiesFormat Kindle/Paperback

The Machine Learning for Dummies book aims to make the readers familiar with the basic concepts and theories pertaining to machine learning in an easy way. Also, the book focuses on the practical, real-world applications of machine learning.

The machine learning book from John Paul Mueller and Luca Massaron uses Python and R code to demonstrate how to train machines to find patterns and analyze results. The book also explains how ml facilitates email filters, fraud detection, internet ads, web searches, etc.

Topics covered

You can buy this book here.

Author John D. Kelleher, Brian Mac Namee, and Aoife DArcyLatest Edition FirstPublisher The MIT PressFormat Hardcover/Kindle

Predictive analytics makes use of an array of statistical techniques that helps in analyzing the past and current events to make future predictions based on the same. The Fundamentals of Machine Learning for Predictive Data Analytics book dives into the basics of machine learning required to do better predictive data analytics.

Obviously, you need to have at least a sound understanding of the basics of predictive data analytics to benefit from the machine learning book. Each machine learning concept explained in the machine learning book comes with suitable algorithms, models, and well-explained examples.

Topics covered

You can buy this book here.

Author Peter HarringtonLatest Edition FirstPublisher Manning PublicationsFormat Paperback

The Machine Learning in Action is yet another opportune machine learning book preferred by a variety of people ranging from undergraduates to professionals. It not only details machine learning techniques but the concepts underlying them as well as in a thoroughly-explained way.

The machine learning book can also act as a walkthrough for developers for writing their own programs meant for acquiring data with the aim of analysis. The Machine Learning in Action book goes in-depth in discussing the algorithms forming the basis of various machine learning techniques. Most examples mentioned in the machine learning book use Python code.

Topics covered

You can buy this book here.

Author Ian H. Witten, Eibe Frank, and Mark A. HallLatest Edition FourthPublisher Morgan KaufmannFormat Kindle/Paperback

Data mining techniques help us discover patterns in large data sets by means of methods that belong to the fields of database systems, machine learning, and statistics. If you need to or plan to learn data mining techniques, in particular, and machine learning, in general then you must pick up the Data Mining: Practical Machine Learning Tools and Techniques book.

The top machine learning book focuses more on the technical aspect of machine learning. It dives deeper into the technical details of machine learning, methods for obtaining data, and using different inputs and outputs for evaluating results.

Topics covered

You can buy this book here.

Author Nishant ShuklaLatest Edition FirstPublisher Manning PublicationsFormat ebook (free)/Paperback

TensorFlow is a symbolic math library, and one of the top data science Python libraries, that is used for machine learning applications, most notably neural networks. The Machine Learning with TensorFlow book offers readers a robust explanation of machine learning concepts and practical coding experience.

The Machine Learning with TensorFlow book explains the ml basics with traditional classification, clustering, and prediction algorithms. The book all dives deeper into deep learning concepts making the readers ready for any kind of machine learning task using the free and open-source TensorFlow library.

Topics covered

You can buy this book here.

Author Aurlien GronLatest Edition SecondPublisher OReilly MediaFormat Kindle/Paperback

The second edition of the Hands-On Machine Learning adds Keras to its content list, alongside Scikit-Learn and TensorFlow. The machine learning book gives an intuitive understanding of the various concepts and tools that you need to develop smart, intelligent systems.

You need programming experience to get started with the Hands-On Machine Learning book. Each chapter in the machine learning book features numerous exercises that will help you apply what youve learned till that time. Post successful reading of the book, one should be able to implement intelligent programs capable of learning from data gained.

Topics covered

You can buy this book here.

Author Andreas C. Mller & Sarah GuidoLatest Edition FirstPublisher OReilly MediaFormat Kindle/Paperback

Are you a data scientist proficient in using Python and interested in learning ML? Then the Introduction to Machine Learning with Python: A Guide for Data Scientists is the ideal book for you to pick up and kickstart your machine learning journey.

The Introduction to Machine Learning with Python: A Guide for Data Scientists book will teach you various practical ways of building your very own machine learning solutions.

You will get to know all the important steps for creating robust machine learning applications using Python and Scikit-learn library. Having a good understanding of matplotlib and NumPy libraries will help the learning process even better.

Topics covered

You can buy this book here.

Author Kevin P. MurphyLatest Edition FirstPublisher The MIT PressFormat eTextbook/Hardcover

Full of informal writing and pseudocode for important algorithms, the Machine Learning: A Probabilistic Perspective is a fun machine learning book that flaunts nostalgic color images and practical, real-world examples belonging to various domains like biology, computer vision, robotics, and text processing.

Unlike other machine learning books that are written like a cookbook explaining several heuristic methods, the Machine Learning: A Probabilistic Perspective focuses on a principled model-based approach. It uses graphical models for specifying ml models in a concise, intuitive way.

Topics covered

You can buy this book here.

Author Leonard EddisonLatest Edition FirstPublisher CreateSpace Independent Publishing PlatformFormat Audiobook/Paperback

Read more:
20 Best Machine Learning Books for Beginner & Experts in 2021

Read More..

Machine learning algorithm used for natural language processing

The Transformer is a deep learning model introduced in 2017 that utilizes the mechanism of attention. It is used primarily in the field of natural language processing (NLP)[1], but recent research has also developed its application in other tasks like video understanding.[2]

Like recurrent neural networks (RNNs), Transformers are designed to handle sequential data, such as natural language, for tasks such as translation and text summarization. However, unlike RNNs, Transformers do not require that the sequential data be processed in order. For example, if the input data is a natural language sentence, the Transformer does not need to process the beginning of it before the end. Due to this feature, the Transformer allows for much more parallelization than RNNs and therefore reduced training times.[1]

Transformers have rapidly become the model of choice for NLP problems,[3] replacing older recurrent neural network models such as the long short-term memory (LSTM). Since the Transformer model facilitates more parallelization during training, it has enabled training on larger datasets than was possible before it was introduced. This has led to the development of pretrained systems such as BERT (Bidirectional Encoder Representations from Transformers) and GPT (Generative Pre-trained Transformer), which have been trained with huge general language datasets, such as Wikipedia Corpus and Common Crawl, and can be fine-tuned to specific language tasks.[4][5]

Before the introduction of Transformers, most state-of-the-art NLP systems relied on gated recurrent neural networks (RNNs), such as LSTMs and gated recurrent units (GRUs), with added attention mechanisms. The Transformer built on these attention technologies without using an RNN structure, highlighting the fact that the attention mechanisms alone, without recurrent sequential processing, are powerful enough to achieve the performance of RNNs with attention.

Gated RNNs process tokens sequentially, maintaining a state vector that contains a representation of the data seen after every token. To process the n t h {textstyle n^{th}} token, the model combines the state representing the sentence up to token n 1 {textstyle n-1} with the information of the new token to create a new state, representing the sentence up to token n {textstyle n} . Theoretically, the information from one token can propagate arbitrarily far down the sequence, if at every point the state continues to encode information about the token. But in practice this mechanism is imperfect: due in part to the vanishing gradient problem, the model's state at the end of a long sentence often does not contain precise, extractable information about early tokens.

This problem was addressed by the introduction of attention mechanisms. Attention mechanisms let a model directly look at, and draw from, the state at any earlier point in the sentence. The attention layer can access all previous states and weighs them according to some learned measure of relevancy to the current token, providing sharper information about far-away relevant tokens. A clear example of the utility of attention is in translation. In an English-to-French translation system, the first word of the French output most probably depends heavily on the beginning of the English input. However, in a classic encoder-decoder LSTM model, in order to produce the first word of the French output the model is only given the state vector of the last English word. Theoretically, this vector can encode information about the whole English sentence, giving the model all necessary knowledge, but in practice this information is often not well preserved. If an attention mechanism is introduced, the model can instead learn to attend to the states of early English tokens when producing the beginning of the French output, giving it a much better concept of what it is translating.

When added to RNNs, attention mechanisms led to large gains in performance. The introduction of the Transformer brought to light the fact that attention mechanisms were powerful in themselves, and that sequential recurrent processing of data was not necessary for achieving the performance gains of RNNs with attention. The Transformer uses an attention mechanism without being an RNN, processing all tokens at the same time and calculating attention weights between them. The fact that Transformers do not rely on sequential processing, and lend themselves very easily to parallelization, allows Transformers to be trained more efficiently on larger datasets.

Like the models invented before it, the Transformer is an encoder-decoder architecture. The encoder consists of a set of encoding layers that processes the input iteratively one layer after another and the decoder consists of a set of decoding layers that does the same thing to the output of the encoder.

The function of each encoder layer is to process its input to generate encodings, containing information about which parts of the inputs are relevant to each other. It passes its set of encodings to the next encoder layer as inputs. Each decoder layer does the opposite, taking all the encodings and processes them, using their incorporated contextual information to generate an output sequence.[6] To achieve this, each encoder and decoder layer makes use of an attention mechanism, which for each input, weighs the relevance of every other input and draws information from them accordingly to produce the output.[7] Each decoder layer also has an additional attention mechanism which draws information from the outputs of previous decoders, before the decoder layer draws information from the encodings. Both the encoder and decoder layers have a feed-forward neural network for additional processing of the outputs, and contain residual connections and layer normalization steps.[7]

The basic building blocks of the Transformer are scaled dot-product attention units. When a sentence is passed into a Transformer model, attention weights are calculated between every token simultaneously. The attention unit produces embeddings for every token in context that contain information not only about the token itself, but also a weighted combination of other relevant tokens weighted by the attention weights.

Concretely, for each attention unit the Transformer model learns three weight matrices; the query weights W Q {displaystyle W_{Q}} , the key weights W K {displaystyle W_{K}} , and the value weights W V {displaystyle W_{V}} . For each token i {displaystyle i} , the input word embedding x i {displaystyle x_{i}} is multiplied with each of the three weight matrices to produce a query vector q i = x i W Q {displaystyle q_{i}=x_{i}W_{Q}} , a key vector k i = x i W K {displaystyle k_{i}=x_{i}W_{K}} , and a value vector v i = x i W V {displaystyle v_{i}=x_{i}W_{V}} . Attention weights are calculated using the query and key vectors: the attention weight a i j {displaystyle a_{ij}} from token i {displaystyle i} to token j {displaystyle j} is the dot product between q i {displaystyle q_{i}} and k j {displaystyle k_{j}} . The attention weights are divided by the square root of the dimension of the key vectors, d k {displaystyle {sqrt {d_{k}}}} , which stabilizes gradients during training, and passed through a softmax which normalizes the weights to sum to 1 {displaystyle 1} . The fact that W Q {displaystyle W_{Q}} and W K {displaystyle W_{K}} are different matrices allows attention to be non-symmetric: if token i {displaystyle i} attends to token j {displaystyle j} (i.e. q i k j {displaystyle q_{i}cdot k_{j}} is large), this does not necessarily mean that token j {displaystyle j} will attend to token i {displaystyle i} (i.e. q j k i {displaystyle q_{j}cdot k_{i}} is large). The output of the attention unit for token i {displaystyle i} is the weighted sum of the value vectors of all tokens, weighted by a i j {displaystyle a_{ij}} , the attention from token i {displaystyle i} to each token.

The attention calculation for all tokens can be expressed as one large matrix calculation, which is useful for training due to computational matrix operation optimizations which make matrix operations fast to compute. The matrices Q {displaystyle Q} , K {displaystyle K} and V {displaystyle V} are defined as the matrices where the i {displaystyle i} th rows are vectors q i {displaystyle q_{i}} , k i {displaystyle k_{i}} , and v i {displaystyle v_{i}} respectively.

Attention ( Q , K , V ) = softmax ( Q K T d k ) V {displaystyle {begin{aligned}{text{Attention}}(Q,K,V)={text{softmax}}left({frac {QK^{mathrm {T} }}{sqrt {d_{k}}}}right)Vend{aligned}}}

One set of ( W Q , W K , W V ) {displaystyle left(W_{Q},W_{K},W_{V}right)} matrices is called an attention head, and each layer in a Transformer model has multiple attention heads. While one attention head attends to the tokens that are relevant to each token, with multiple attention heads the model can learn to do this for different definitions of "relevance". Research has shown that many attention heads in Transformers encode relevance relations that are interpretable by humans. For example there are attention heads that, for every token, attend mostly to the next word, or attention heads that mainly attend from verbs to their direct objects.[8] Since Transformer models have multiple attention heads, they can perform computations in parallel, which allows for a fast processing of the input sequence. The multiple outputs for the multi-head attention layer are concatenated to pass into the feed-forward neural network layers.

Each encoder consists of two major components: a self-attention mechanism and a feed-forward neural network. The self-attention mechanism takes in a set of input encodings from the previous encoder and weighs their relevance to each other to generate a set of output encodings. The feed-forward neural network then further processes each output encoding individually. These output encodings are finally passed to the next encoder as its input, as well as the decoders.

The first encoder takes positional information and embeddings of the input sequence as its input, rather than encodings. The positional information is necessary for the Transformer to make use of the order of the sequence, because no other part of the Transformer makes use of this.[1]

Each decoder consists of three major components: a self-attention mechanism, an attention mechanism over the encodings, and a feed-forward neural network. The decoder functions in a similar fashion to the encoder, but an additional attention mechanism is inserted which instead draws relevant information from the encodings generated by the encoders.[1][7]

Like the first encoder, the first decoder takes positional information and embeddings of the output sequence as its input, rather than encodings. Since the transformer should not use the current or future output to predict an output though, the output sequence must be partially masked to prevent this reverse information flow.[1] The last decoder is followed by a final linear transformation and softmax layer, to produce the output probabilities over the vocabulary.

Below is pseudo code for an implementation of the Transformer variant known as the "vanilla" transformer:

Training Transformer-based architectures can be very expensive, especially for long sentences.[9] Alternative architectures include the Reformer, which reduces the computational load from O ( N 2 ) {displaystyle O(N^{2})} to O ( N ln N ) {displaystyle O(Nln N)} , where N {displaystyle N} is the length of the sequence. This is done using locality-sensitive hashing and reversible layers.[10][11]

A benchmark for comparing different transformer architectures was introduced in late 2020.[12]

Transformers typically undergo semi-supervised learning involving unsupervised pretraining followed by supervised fine-tuning. Pretraining is typically done on a much larger dataset than fine-tuning, due to the restricted availability of labeled training data. Tasks for pretraining and fine-tuning commonly include:

The Transformer finds most of its applications in the field of natural language processing (NLP), for example the tasks of machine translation and time series prediction.[14] Many pretrained models such as GPT-2, GPT-3, BERT, XLNet, and RoBERTa demonstrate the ability of Transformers to perform a wide variety of such NLP-related tasks, and have the potential to find real-world applications.[4][5][15] These may include:

In 2020, it was shown that the transformer architecture, more specifically GPT-2, could be fine-tuned to play chess.[20] Transformers have also been applied to image processing with results showing their ability to compete with convolutional neural networks.[21][22]

The Transformer model has been implemented in major deep learning frameworks such as TensorFlow and PyTorch.

Transformers is a library produced by Hugging Face which supplies Transformer-based architectures and pretrained models.[3] The library is free software and available on GitHub.[3] Its models are available both in PyTorch and TensorFlow format.[3]

See original here:
Transformer (machine learning model) - Wikipedia

Read More..

Machine learning for enterprise use is exploding. From improving customer experience to developing products, there's almost no area of the modern business untouched by machine learning.

Machine learning is a pathway to creating artificial intelligence, which in turn is one of the primary drivers of machine learning use in the enterprise. There is some disagreement over the exact nature of the relationship between AI and machine learning. Some see machine learning as a subfield of AI, while others view AI essentially as a subfield of machine learning. In general, AI aims to replicate some aspect of human perception or decision-making, whereas machine learning can be used to enhance or automate virtually any task, not just ones related to human cognition. However you view them, the two concepts are closely linked, and they are feeding off each other's popularity.

The practice of machine learning involves taking data, examining it for patterns and developing some sort of prediction about future outcomes. By feeding an algorithm more data over time, data scientists can sharpen the machine learning model's predictions. From this basic concept, a number of different types of machine learning have developed:

From these four main types of machine learning, enterprises have developed an impressive array of techniques and applications. Everything from relatively simple sales forecasting to today's most cutting-edge AI tools run on machine learning models. This guide to machine learning in the enterprise explores the variety of use cases for machine learning, the challenges to adoption, how to implement machine learning technologies and much more.

Machine learning for enterprise use is accelerating, and not just at the periphery. Increasingly, businesses are putting machine learning applications at the center of their business models. The technology has enabled businesses to perform tasks at a scale previously unachievable, not only generating efficiencies for companies but also new business opportunities, as technology writer Mary Pratt explained in "10 common uses for machine learning in business." The growing use of machine learning in mission-critical business processes is reflected in the range of use cases where it plays an integral role. The following are examples:

These are just some examples, but there are countless more. Any business process that either produces or uses large amounts of data -- particularly structured, labeled data -- is ripe for automation that uses machine learning. Enterprises across all industries have learned this and are working to implement machine learning methods throughout their processes.

It's not hard to see why machine learning has entered so many situations. Enterprises that have adopted machine learning are solving business problems and reaping value from this AI technique. Here are six business benefits:

The question is no longer whether to use machine learning, it's how to operationalize machine learning in ways that return optimal results. That's where things get tricky.

Machine learning is a complicated technology that requires substantial expertise. Unlike some other technology domains, where software is mostly plug and play, machine learning forces the user to think about why they are using it, who is building the tools, what their assumptions are and how the technology is being applied. There are few other technologies that have so many potential points of failure.

The wrong use case is the downfall of many machine learning applications. Sometimes enterprises lead with the technology, looking for ways to implement machine learning, rather than allowing the problem to dictate the solution. When machine learning is shoehorned into a use case, it often fails to deliver results.

The wrong data dooms machine learning models faster than anything. Data is the lifeblood of machine learning. Models only know what they've been shown, so when the data they train on is inaccurate, unorganized or biased in some way, the model's output will be faulty.

Bias frequently hampers machine learning implementations. The many types of bias that can undermine machine implementations generally fall into the two categories. One type happens when data collected to train the algorithm simply doesn't reflect the real world. The data set is inaccurate, incomplete or not diverse enough. Another type of bias stems from the methods used to sample, aggregate, filter and enhance that data. In both cases, the errors can stem from the biases of the data scientists overseeing the training and result in models that are inaccurate and, worse, unfairly affect specific populations of people. In his article "6 ways to reduce different types of bias in machine learning," analyst Ron Schmelzer explained the types of biases that can derail machine learning projects and how to mitigate them.

Black box functionality is one reason why bias is so prevalent in machine learning. Many types of machine learning algorithms -- particularly unsupervised algorithms -- operate in ways that are opaque, or as a "black box," to the developer. A data scientist feeds the algorithm data, the algorithm makes observations of correlations and then produces some sort of output based on these observations. But most models can't explain to the data scientist why they produce the outputs they do. This makes it extremely difficult to detect instances of bias or other failures of the model.

Technical complexity is one of the biggest challenges to enterprise use of machine learning. The basic concept of feeding training data to an algorithm and letting it learn the characteristics of the data set may sound simple enough. But there is a lot of technical complexity under the hood. Algorithms are built around advanced mathematical concepts, and the code that algorithms run on can be difficult to learn. Not all businesses have the technical expertise in house needed to develop effective machine learning applications.

Lack of generalizability prevents machine learning from scaling to new use cases in most enterprises. Machine learning applications only know what they've been explicitly trained on. This means a model can't take something it learned about one area and apply it to another, the way a human would be able to. Algorithms need to be trained from scratch for every new use case.

To learn more about machine learning, here is a list of nine books ranging from a concise introduction for beginners to advanced texts on cutting-edge techniques by AI's leading experts.

Implementing machine learning is a multistep process requiring input from many types of experts. Here is an outline of the process in six steps.

The management and maintenance of machine learning applications in the enterprise is one area that's sometimes given short shrift, but it can be what makes or breaks use cases.

The basic functionality of machine learning depends on models learning trends -- such as customer behavior, stock performance and inventory demand -- and projecting them to the future to inform decisions. However, underlying trends are constantly shifting, sometimes slightly, sometimes substantially. This is called concept drift, and if data scientists don't account for it in their models, the model's projections will eventually be off base.

The way to correct for this is to never view models in production as finished. They demand a constant state of verification, retraining and reworking to ensure they continue to deliver results.

Machine learning operations, or MLOps, is an emerging concept aimed at actively managing this lifecycle. Rather than an ad hoc approach to verifying and retraining when appropriate, MLOps tools put each model on a schedule for development, deployment, verification and retraining. It seeks to standardize these processes, a practice that's becoming more important as enterprises make machine learning a core component of their operations.

When we look to the future of machine learning, one overarching trend predominates. Enterprise adoption will continue to increase, bringing the technology from cutting edge to mainstream.

The trend is already well underway.

A 2019 survey from analyst firm Gartner found that 37% of enterprises have adopted some form of artificial intelligence. That's up from 10% in 2015. At its current trajectory, machine learning is on a path to become a ubiquitous technology in the next few years. In its ranking of the top 10 data and analytics trends for 2020, the analyst firm named "smarter, faster and more responsible AI" as the year's top trend. The report, noting the vital importance of machine learning and other AI techniques in providing insight into the global coronavirus pandemic, predicted that by 2024, 75% of organizations will have shifted from piloting to operationalizing AI. As a result of high rates of adoption of machine learning in the enterprise, the market for machine learning tools is growing rapidly. The analyst firm Research and Markets predicted that the machine learning market will grow to $8.8 billion by 2022, from $1.4 billion in 2017.

The reasons for this are clear. Today's most successful companies, like Amazon, Google and Uber, put machine learning applications at the center of their business models. Rather than viewing machine learning as a nice-to-have technology, industry-leading enterprises are using machine learning and AI technologies as critical to maintaining their competitive edge, as technology writer George Lawton explored in "Learn the business value of AI's various techniques."

Advances in deep learning -- a type of machine learning based on neural networks -- have played a huge role in bringing AI to the fore in the enterprise. Neural networks are relatively common in enterprise applications today. These advanced deep learning techniques enable models to do everything from recognize objects in images to create natural language text for product descriptions and other applications. Today, there are a number of different types of neural networks, which are designed to perform specific jobs. As technology writer David Petersson explained in "CNNs vs. RNNs: How they differ and where they overlap," understanding the uniqueness of different types of algorithms is key to getting the most out of them.

It is now viewed as inevitable that a large amount of knowledge work will be automated. Even some creative fields are being infiltrated by machine learning-driven AI applications. This is raising questions about the future of work. In a world where machines are able to manage customer relations, detect cancer in medical images, conduct legal reviews, drive shipping containers across the country and produce creative assets, what is the role of human workers? Proponents of AI say automation will free people up to pursue more creative activities by eliminating rote tasks. But others worry that an incessant drive for automation will leave little room for human workers.

Enterprises looking to deploy machine learning have no shortage of options. The machine learning space features strong competition between open source tools and software built and supported by traditional vendors. Regardless of whether an enterprise chooses machine learning software from a vendor or open source tool, it is common for applications to be hosted in the cloud computing environments and delivered as a service. There are more vendors and platforms than one article could name, but the following list gives a high-level overview of offerings from some of the bigger players in the field.

A more exhaustive list of vendor offerings can be found in this expert overview of machine learning platforms.

In general, most enterprise machine learning users consider open source tools to be more innovative and powerful. However, there is still a strong case for proprietary tools, as vendors offer training and support that is generally absent from open source offerings. Many of today's vendor tools support use of open source libraries, allowing users to have the best of both worlds.

Read more from the original source:
In-Depth Guide to Machine Learning in the Enterprise

Read More..

If you are interested in pursuing a career in AI and don't know where to start, here's your go-to guide for the best programming languages and skills to learn, interview questions, salaries, and more.

Machine learning engineers--i.e., advanced programmers who develop artificial intelligence (AI) machines and systems that can learn and apply knowledge--are in high demand, as more companies adopt these technologies. These professionals perform sophisticated programming, and work with complex data sets and algorithms to train intelligent systems.

While many fear that AI will soon replace jobs, at this phase in the technology's development, it is still creating positions like machine learning engineers, as companies need highly-skilled workers to develop and maintain a wide range of applications.

To help those interested in the field better understand how to break into a career in machine learning, we compiled the most important details and resources. This guide on how to become a machine learning engineer will be updated on a regular basis.

SEE: Managing AI and ML in the enterprise (ZDNet special report) | Download the report as a PDF (TechRepublic)

According to TechRepublic writers Hope Reese and Brandon Vigliarolo, machine learning is a branch of AI that gives computer systems the ability to automatically learn and improve from experience, rather than being explicitly programmed. In machine learning, computers use massive sets of data and apply algorithms to train on and make predictions.

Machine learning systems are able to rapidly apply knowledge and training from large data sets to perform facial recognition, speech recognition, object recognition, translation, and many other tasks.

Additional resources

Demand for AI talent, including machine learning engineers, is exploding: Between June 2015 and June 2018, the number of job postings with "AI" or "machine learning" increased by nearly 100%, according to a report from job search site Indeed. The percent of searches for these terms on Indeed also increased by 182% in that time frame, the report found.

"There is a growing need by employers for AI talent," Raj Mukherjee, senior vice president of product at Indeed, told TechRepublic. "As companies continue to adopt solutions or develop their own in-house it is likely that demand by employers for these skills will continue to rise."

SEE: IT jobs 2018: Hiring priorities, growth areas, and strategies to fill open roles (Tech Pro Research)

In terms of specific positions, 94% of job postings that contained AI or machine learning terminology were for machine learning engineers, the report found. And 41% of machine learning engineer positions were still open after 60 days.

"Software is eating the world and machine learning is eating software," Vitaly Gordon, vice president of data science and software engineering for Salesforce Einstein, told TechRepublic. "Machine learning engineering is a discipline that requires production grade coding, PhD level machine learning, and the business acumen of a product manager. Finding such rare people can uplift a company from a follower into a leader in their space, and everyone is looking for them."

Additional resources

Machine learning engineers can take a number of different career paths. Here are a few roles in the field, and the skills they require, according to Udacity.

Additional resources

Python and R are the most popular programming languages for machine learning, data science, and analytics, according to a KDnuggets survey. Python had a 66% share of voters who used the tool in 2018--an increase of 11% from 2017. Meanwhile, R had a 49% share in 2018, down 14% from 2017.

An IBM report ranked Python, Java, and R as the top languages for machine learning engineers, followed by C++, C, JavaScript, Scala, and Julia.

SEE: All of TechRepublic's cheat sheets and smart person's guides

When developing machine learning applications, the training and operational phases for algorithms are different, as reported by our sister site ZDNet. Therefore, some people use one language for the training phase and another one for the operational phase.

"For 'ordinary machine learning,' it does not matter what language you use," Luiz Eduardo Le Masson, data science leader at Stone Co., told ZDNet. "But when you need to have real online learning algorithms and inferences in realtime for millions of simultaneous clusters and respond in less than 500 ms, the topic does not only involve languages, but architecture, design, flow control, fault tolerance, resilience."

Additional resources

Generally, machine learning engineers must be skilled in computer science and programming, mathematics and statistics, data science, deep learning, and problem solving. Here is a breakdown of some of the skills needed, according to Udacity.

Additional resources

Machine learning engineers in the US earn an average salary of $134,449, according to data from Indeed. In terms of AI-related jobs, it comes in third place for salary, after director of analytics ($140,837) and principal scientist ($138,271).

Additional resources

New York City has the highest concentration of AI jobs, with nearly 12% of all AI job postings found there, according to Indeed. New York also has the highest concentrations of data engineer, data scientist, and director of analytics job postings of any US metro area, potentially supporting the media, fashion, and banking industry centers located there, Indeed found.

Following New York City in AI job concentration is San Francisco (10%), San Jose, CA (9%), Washington, DC (8%), Boston (6%), and Seattle (6%). San Jose has the most postings for machine learning engineers in particular, along with algorithm engineers, computer vision engineers, and research engineers.

Additional resources

Those applying for machine learning jobs can expect a number of different types of questions during an interview, testing their skills in mathematics and statistics, data science, deep learning, programming, and problem solving.

Some questions that a machine learning engineer can expect to be asked during an interview include:

It's also important for the job applicant to arrive at the interview with questions for the hiring manager, Dave Castillo, managing vice president of machine learning at Capital One told TechRepublic.

"An interview is a two-way conversation," Castillo said. "Just as important as the questions that we ask are the questions that candidates ask us. We want to ensure that not only is the candidate the right choice for the company, but the company is the right choice for the candidate."

Additional resources

There are different paths into a career as a machine learning engineer. A good place to start is by learning a programming language like Python, R, or Java. For machine learning specifics, a number of Massive Open Online Courses (MOOCs), online programs, and certifications are available, including classes on Coursera and edX, and a nanodegree from Udacity.

You can also gain practical experience through doing real projects on real data, on sites like Kaggle. Joining local organizations such as meetups or hackathons to learn from others in the field can also help.

Additional resources

Discover the secrets to IT leadership success with these tips on project management, budgets, and dealing with day-to-day challenges. Delivered Tuesdays and Thursdays

Image: iStockphoto/Gorodenkoff Productions OU

See original here:
How to become a machine learning engineer: A cheat sheet ...

Read More..