LeoGlossary: Machine Learning

avatar
(Edited)

How to get a Hive Account


Machine learning (ML) is the sub-category of artificial intelligence (AI) that builds models of algorithms to identify patterns and relationships in data.

The three sub-categories are usually recognized as:

These are actually listed in sub-category (or field) order. Neural networks are a sub-field of machine while deep learning is the same for neural networks.

Machine learning focuses on the use of data and algorithms to imitate the way that humans learn, gradually improving its accuracy.

Machine refers to software, or a computer program, while the learning describes the algorithms becoming more accurate.

Machine learning approaches have been applied to:

  • large language models
  • computer vision
  • speech recognition
  • email filtering
  • agriculture
  • medicine

This is used when the developing of algorithms by human programmers would be cost prohibitive.

Machine learning is an important component of the growing field of data science. Through the use of statistical methods, algorithms are trained to make classifications or predictions, and to uncover key insights in data mining projects. These insights subsequently drive decision making within applications and businesses, ideally impacting key growth metrics.

History of Machine Learning

A Journey Through Time: The History of Machine Learning

Machine learning, the fascinating field where computers learn and improve without explicit programming, boasts a rich and dynamic history filled with pivotal moments and intriguing characters. Let's go through time to explore its key milestones:

Early Seeds (1940s-1960s):

  • 1943: The seeds are sown with Walter Pitts and Warren McCulloch's "threshold logic" model, mimicking the brain's neural networks.
  • 1950s: Alan Turing's Turing Test sets the benchmark for machine intelligence, while Frank Rosenblatt invents the Perceptron, a single-layer neural network for pattern recognition.
  • 1960s: The Perceptron encounters limitations, leading to a period of skepticism and reduced research in neural networks, often referred to as the "AI winter."

Resurgence and Growth (1970s-1990s):

  • 1970s: Hope rekindles with Geoffrey Hinton's backpropagation algorithm, enabling the training of multi-layer neural networks and overcoming limitations of early models.
  • 1980s: The term "deep learning" is arguably coined by Rina Dechter, and research in convolutional neural networks (CNNs) by Yann LeCun and recurrent neural networks (RNNs) by Sepp Hochreiter and Jürgen Schmidhuber lays the foundation for future advancements.
  • 1990s: CNNs revolutionize computer vision by automating feature extraction from images, while RNNs learn long-term dependencies in sequential data, powering applications like language translation.

Breakthroughs and Boom (2000s-Present):

  • 2006: Geoffrey Hinton reintroduces the term "deep learning," marking a shift in perception and renewed enthusiasm.
  • 2012: AlexNet, a deep CNN, wins the ImageNet competition, showcasing the remarkable capabilities of deep learning and sparking a surge in research and applications.
  • 2010s-Present: Deep learning explodes across various fields like computer vision, natural language processing (NLP), and speech recognition. Advancements in hardware like GPUs accelerate training and development.

Key Turning Points:

  • Backpropagation: Unlocked the potential of multi-layer networks, pushing the boundaries of learning complexity.
  • CNNs: Revolutionized computer vision by automating feature extraction and enabling machines to "see" the world like humans.
  • RNNs: Learned long-term dependencies in sequential data, opening doors for applications like language translation and speech recognition.
  • Availability of data and powerful hardware: Fueled the exponential growth of deep learning research and applications, making it a ubiquitous technology.

The Future Unfolds:

Machine learning's history is still being written, with constant advancements and new challenges emerging. From healthcare, robotics and personalized education, the potential applications seem limitless. As we strive to develop responsible and ethical AI, deep learning remains a powerful tool shaping the future of technology and impacting our lives in profound ways.

General:

Posted Using InLeo Alpha



0
0
0.000
0 comments