Everand Logo

Welcome to Everand!

  • Discover millions of ebooks, audiobooks, and so much more with a free trial

    From $11.99/month after trial. Cancel anytime.

    GROKKING ALGORITHMS: Simple and Effective Methods to Grokking Deep Learning and Machine Learning
    GROKKING ALGORITHMS: Simple and Effective Methods to Grokking Deep Learning and Machine Learning
    GROKKING ALGORITHMS: Simple and Effective Methods to Grokking Deep Learning and Machine Learning
    Ebook210 pages2 hours

    GROKKING ALGORITHMS: Simple and Effective Methods to Grokking Deep Learning and Machine Learning

    By Eric Schmidt

    ()

    Read preview

    About this ebook

    In the past decade, artificial intelligence has been making waves. From self-driving cars to Siri to Alexa, artificial intelligence is everywhere. But what exactly is it?


    The term "artificial intelligence" was coined in 1956 by John McCarthy, a computer scientist at Dartmouth College. His work was based on the idea that comput

    LanguageEnglish
    PublisherEric Schmidt
    Release dateJul 26, 2023
    ISBN9781088225363
    GROKKING ALGORITHMS: Simple and Effective Methods to Grokking Deep Learning and Machine Learning
    Author

    Eric Schmidt

    Eric Schmidt served as Google CEO and chairman from 2001 until 2011, Google executive chairman from 2011 to 2015, and Alphabet executive chairman from 2015 to 2018.

    Read more from Eric Schmidt

    Related to GROKKING ALGORITHMS

    Related ebooks

    Programming For You

    View More
    View More

    Related podcast episodes

    Related articles

    • UNLIMITED

      Tensor Flow 101

      Jan 27, 2020

      Tensor Flow 101
      4 min read

    • UNLIMITED

      Gödel, Escher, Bach, and AI

      Jul 8, 2023

      Gödel, Escher, Bach, and AI
      10 min read

    • UNLIMITED

      Create A RESTful Server In Go

      Oct 19, 2021

      Create A RESTful Server In Go
      8 min read

    • UNLIMITED

      RISC-V on Ubuntu

      May 2, 2023

      RISC-V on Ubuntu
      1 min read

    • UNLIMITED

      Top 10 Programming Languages

      Jan 5, 2023

      Top 10 Programming Languages
      8 min read

    • UNLIMITED

      Access Your Mac Anywhere

      Nov 8, 2022

      Access Your Mac Anywhere
      2 min read

    • UNLIMITED

      The Art Of AI

      Feb 7, 2023

      “Everywhere you look, people are discussing the AI chatbot ChatGPT. It’s been applied to write email campaigns, code and articles. It’s been banned in some schools as students use it to write essays. Singer Nick Cave called ChatGPT’s output in the st

      1 min read

    • UNLIMITED

      AWS vs Azure

      Aug 22, 2023

      AWS vs Azure
      9 min read

    • UNLIMITED

      Build The Kernel

      Mar 8, 2022

      Build The Kernel
      1 min read

    • UNLIMITED

      A.i. Coding

      Aug 22, 2023

      A.i. Coding
      16 min read

    • UNLIMITED

      AI: Problems and Promises

      May 11, 2023

      Will artificial intelligence (AI) replace my job? This is a question that lingers in the minds of many these days. Similar concerns surfaced a couple of years ago when said technology narrowly defeated a champion of the game Go. At present, the appli

      1 min read

    • UNLIMITED

      Stay Safe Online!

      Jan 9, 2024

      Stay Safe Online!
      19 min read

    • UNLIMITED

      How AI Amplifies Human Competencies

      Jan 1, 2020

      Not everyone is on board with AI and machine learning. None other than Elon Musk has suggested that it might lead to World War III. What is your take on the situation? You’re absolutely right that there is a huge amount of alarmism and ‘automation an

      5 min read

    • UNLIMITED

      ChatGPT Is Already Obsolete

      May 19, 2023

      ChatGPT Is Already Obsolete
      5 min read

    Related categories

    Reviews for GROKKING ALGORITHMS

    0 ratings

    0 ratings0 reviews

    What did you think?

    Tap to rate

    Review must be at least 10 words

      Book preview

      GROKKING ALGORITHMS - Eric Schmidt

      Introduction

      In the past decade, artificial intelligence has been making waves. From self-driving cars to Siri to Alexa, artificial intelligence is everywhere. But what exactly is it?

      The term artificial intelligence was coined in 1956 by John McCarthy, a computer scientist at Dartmouth College. His work was based on the idea that computers could be made to exhibit intelligent behavior if they were programmed correctly.

      Since then, many people have taken up the mantle of creating intelligent machines—from engineers and computer scientists like Alan Turing and John von Neumann to psychologists like B. F. Skinner and Ivan Pavlov. And today, dozens of different types of artificial intelligence algorithms are being used by companies worldwide: deep learning algorithms, machine learning algorithms… even genetic algorithms!

      There are a lot of algorithms out there. Some of them are old, and some of them are new. But if you're reading this book, chances are you've heard something about machine learning, maybe even deep learning.

      The world of computer science is a vast and fascinating place. It's also constantly changing, with new techniques emerging every day to help us understand data and make better decisions about how we use it.

      One of these techniques is called deep learning. Deep learning allows computers to learn from huge amounts of data, and it's been used to help computers recognize images, understand language, and even drive cars.

      This book will teach you all about the Grokking algorithms behind machine learning and deep learning: what they are, how they work, and where they come from. We'll start with the basics of what an algorithm is in general—what it does, how it works—and then move into more specific details about what makes these particular algorithms so powerful.

      Grokking Algorithms is a deep dive into the world of machine learning and artificial intelligence. You'll learn what makes these systems tick and how they work.

      Chapter 1

      Introduction to Deep Learning

      Deep learning is one of the most powerful and important tools in artificial intelligence. Its potential lies in automating mundane, time-consuming tasks, freeing up valuable time for more important tasks. That said, it can also be fun, and when you simulate human intelligence, you can learn an awful lot about what it means to be human. It won’t be easy to learn, but it is rewarding.

      Why This Book?

      Do you want to learn about deep learning? Are you interested in artificial intelligence but don't know where to start? Grokking Deep Learning is an excellent choice for anyone who wants to learn about both topics.

      Deep learning is a new and powerful tool for the incremental automation of intelligence. In other words, it's a way to program computers to do what they do best: learn from examples and make predictions based on those examples.

      Deep learning can generally be described as an approach to machine learning that uses multiple layers of artificial neural networks (ANNs) to process information. A typical ANN consists of several layers (hence deep), each containing thousands or millions of processing units called neurons connected in complex patterns. These multilayered networks are trained by feeding them many examples and adjusting their connections until the network reliably produces the correct answer for new inputs it hasn't seen before—a technique known as backpropagation training.

      Deep learning is a new and powerful tool for the incremental automation of intelligence. In other words, it's a way to program computers to do what they do best: learn from examples and make predictions based on those examples. Deep learning can generally be described as an approach to machine learning that uses multiple layers of artificial neural networks (ANNs) to process information. A typical ANN consists of several layers (hence deep), each containing thousands or millions of processing

      units called neurons connected in complex patterns. These multilayered networks are trained by feeding them many examples and adjusting their connections until the network reliably produces the correct answer for new inputs it hasn't seen before—a technique known as backpropagation training. Deep learning is machine learning (ML), but not all forms of ML are deep learning. For example, decision trees and linear regression models, which can be very accurate predictors, aren't part of the deep learning family. While predictive accuracy and ease of use are essential measures, deep learning solves complex problems in which data comes from multiple sources (such as images, text, and audio) or when there's a large amount of data available to train on. Deep learning models can also be used to learn new features during training. For example, researchers have used neural networks to learn how to recognize cats based only on examples of cat photos.

      Deep learning differs from other forms of AI in that it doesn't require much human intervention to tell the computer system what to do or how to perform a task. In addition, it's different from most other types of machine learning because it relies on multiple layers of computing units called neurons that are connected in complex patterns. These multilayered networks are trained by feeding them many examples and adjusting their connections until the network reliably produces the correct answer for new inputs it hasn't seen.

      Skilled labor jobs require a long-term investment in education and training (i.e., college degrees). Examples include doctors, lawyers, accountants, and financial analysts. These roles involve higher levels of analysis and problem solving than others; they require a deep understanding of some discipline or field to solve problems. The current trend in these roles is towards deeper specialization within existing disciplines - even as technologies develop new ways of solving problems once considered too difficult for automation (e.g., automated legal discovery).

      It is often suggested that deep learning has the potential to automate skilled labor - a trend that will have implications for employment, education, and productivity. However, this claim is not always supported by evidence. For example, while deep learning can automate tasks like image classification (e.g., recognizing objects in pictures), it cannot replace human judgment or decision-making.

      This is because humans can make decisions based on multiple factors, while machine learning systems can only learn one thing simultaneously. So, for instance, if you want to build a system that identifies cats in images (i.e., what we call an image classifier), the dataset must contain images of cats labeled as such - not just any picture with felines in it.

      In other words, while deep learning has the potential to reduce the need for skilled labor in some fields, it is not clear that this will be an overall trend. Indeed, as automation and artificial intelligence become more common in our lives (from self-driving cars to personal robots), humans may need even more education to adapt quickly enough.

      Deep learning is a creative process. You'll learn a lot about being human by trying to simulate intelligence and creativity. You'll learn a lot about being intelligent in making computers think as humans do and vice versa. This process involves both connectionism and symbolic reasoning. And you'll learn how to program complex systems in Python, which will teach you about programming anything in any language—and maybe even about learning languages in general!

      You'll also need to know a bit about mathematics and programming, but you can pick up those skills along the way. First, you'll get your hands dirty by implementing real deep learning algorithms in Python—and then you'll turn them loose on real data problems!

      Is This Hard to Learn?

      If you've ever learned a new programming language or technology, then you know that it's not just about memorizing syntax and functions. Instead, it understands how and why those things work together to solve problems.

      Can you learn deep learning in a week? Probably not. Will it take months? Maybe, if you're starting from scratch. But if you have some background in machine learning and are curious about what deep learning can add to your repertoire of tools, then I think the time investment is worth it!

      This is simple: Deep learning has numerous applications for improving products across many industries (from natural language processing to image recognition). So, if there's one skill set that every data scientist should be familiar with right now, it's deep learning.

      Developing a good intuition for how to solve problems with deep learning can take time. There are many different ways to approach the same problem, and it takes time to learn about all of the tools and techniques available in deep learning. Perhaps most importantly, it takes time to learn about your data - what features might be important and which ones can be ignored? What kinds of preprocessing steps should you take? How should you partition your training/test sets? These questions are important when building a deep learning model and can be hard to answer. So, my advice is this: If you're starting with no experience in machine learning or deep learning, it might not be the best idea to jump into deep learning without first understanding how traditional machine learning algorithms work. But if you have some background knowledge about these topics (for example, linear regression), then you should go for it!

      How Long before You Can Have Some Fun?

      You can get started with deep learning (DL) with a simple model and dataset, but there are some important considerations to remember.

      Start small: The first step is often the hardest—how do I get started? If you find yourself wondering this question, don't worry! You can take many paths depending on the type of data or problem you want to solve. When starting, it may be helpful to start with a very simple model that doesn't require much training time or computational power. This will allow for quick experimentation without having too many other variables getting in the way of understanding how everything works together under the hood.

      Add complexity gradually: As your skills grow and knowledge expands over time, it's easy for things like more advanced algorithms or complex datasets to feel intimidating at first glance! However, suppose one keeps working at it consistently. In that case, all sorts of things become possible eventually--even really cool stuff like using machine learning algorithms as part of your daily life (for example, by using apps like Google Maps).

      Chapter 2

      How Do Machines Learn?

      Machine learning is one of the most exciting fields in computer science. Artificial intelligence allows computers to learn from data without being explicitly programmed. Machine learning algorithms can be grouped into two distinct categories: supervised and unsupervised. This chapter will look at standard machine learning techniques—from deep learning to parametric vs. nonparametric methods—and how each works.

      What Is Deep Learning?

      Deep learning comes under machine learning, and machine

      Enjoying the preview?
      Page 1 of 1