https://mklab.gr/ Fri, 31 Jul 2026 04:07:42 +0000 MyBB https://mklab.gr/showthread.php?tid=1479 Fri, 31 Jul 2026 03:12:31 +0300 mklabgr]]> https://mklab.gr/showthread.php?tid=1479

What If We Got AI Right? 
by Eleanor Drage

Summary

Summary of What If We Got AI Right? by Eleanor Drage
In What If We Got AI Right?, AI ethicist Eleanor Drage challenges the dominant narratives that portray artificial intelligence as either humanity’s salvation or an unavoidable disaster, arguing that these extreme visions distract from the real political, social, and ethical choices shaping AI today. 

The book examines how the AI industry is influenced by powerful corporations, technological competition, and cultural ideas such as the pursuit of immortality, while highlighting concrete problems including biased algorithms, surveillance, environmental costs, and the concentration of power in big tech. 

Drage argues that AI is not an independent force beyond human control but a reflection of the values, institutions, and inequalities of the societies that create it. A better AI future requires moving beyond fear and hype toward democratic oversight, ethical responsibility, and new ways of thinking about technology. She proposes bringing perspectives from feminism, reparative justice, and climate politics into AI development so that machines are designed to serve broader human needs rather than narrow economic interests.

Ultimately, the book is a call to reshape the purpose of AI: not to maximize power or profit, but to build technologies that support fairness, inclusion, and a more sustainable future. 


BOOK]]>

What If We Got AI Right? 
by Eleanor Drage

Summary

Summary of What If We Got AI Right? by Eleanor Drage
In What If We Got AI Right?, AI ethicist Eleanor Drage challenges the dominant narratives that portray artificial intelligence as either humanity’s salvation or an unavoidable disaster, arguing that these extreme visions distract from the real political, social, and ethical choices shaping AI today. 

The book examines how the AI industry is influenced by powerful corporations, technological competition, and cultural ideas such as the pursuit of immortality, while highlighting concrete problems including biased algorithms, surveillance, environmental costs, and the concentration of power in big tech. 

Drage argues that AI is not an independent force beyond human control but a reflection of the values, institutions, and inequalities of the societies that create it. A better AI future requires moving beyond fear and hype toward democratic oversight, ethical responsibility, and new ways of thinking about technology. She proposes bringing perspectives from feminism, reparative justice, and climate politics into AI development so that machines are designed to serve broader human needs rather than narrow economic interests.

Ultimately, the book is a call to reshape the purpose of AI: not to maximize power or profit, but to build technologies that support fairness, inclusion, and a more sustainable future. 


BOOK]]> https://mklab.gr/showthread.php?tid=1408 Wed, 29 Jul 2026 04:28:05 +0300 mklabgr]]> https://mklab.gr/showthread.php?tid=1408 How Smart Machines Think 
by Sean Gerrish

Summary

In How Smart Machines Think, former Google engineer Sean Gerrish provides an accessible, non-technical breakdown of the core machine learning algorithms and technical breakthroughs driving modern artificial intelligence. By exploring iconic milestones—such as self-driving cars navigating the DARPA Grand Challenge, Netflix’s recommendation algorithm competition, IBM’s Jeopardy!-winning Watson, and DeepMind's game-playing agents like AlphaGo—Gerrish demystifies how neural networks and reinforcement learning enable machines to perceive, reason, and act in complex environments. 

Through engaging storytelling focused on the researchers behind these developments, the book translates sophisticated concepts into simple terms, offering a clear guide to how everyday AI systems operate and why collaborative human engineering remains essential to their success.


BOOK]]> How Smart Machines Think 
by Sean Gerrish

Summary

In How Smart Machines Think, former Google engineer Sean Gerrish provides an accessible, non-technical breakdown of the core machine learning algorithms and technical breakthroughs driving modern artificial intelligence. By exploring iconic milestones—such as self-driving cars navigating the DARPA Grand Challenge, Netflix’s recommendation algorithm competition, IBM’s Jeopardy!-winning Watson, and DeepMind's game-playing agents like AlphaGo—Gerrish demystifies how neural networks and reinforcement learning enable machines to perceive, reason, and act in complex environments. 

Through engaging storytelling focused on the researchers behind these developments, the book translates sophisticated concepts into simple terms, offering a clear guide to how everyday AI systems operate and why collaborative human engineering remains essential to their success.


BOOK]]> https://mklab.gr/showthread.php?tid=1229 Tue, 21 Jul 2026 02:54:06 +0300 mklabgr]]> https://mklab.gr/showthread.php?tid=1229
AI Under Attack 
BY [Kris Kimmerle; David Okeyode]

Summary

AI Under Attack provides a practical guide to understanding and securing modern artificial intelligence systems against emerging cyber threats. The book explains why AI security is different from traditional cybersecurity, focusing on new risks created by large language models, generative AI, retrieval-augmented generation (RAG), and autonomous AI agents. 
It explores major attack techniques such as prompt injection, jailbreaks, data poisoning, memory and context manipulation, and exploitation of AI tools. The authors present methods for designing safer AI systems through threat modeling, secure architectures, defensive prompting, guardrails, and Zero Trust principles. The guide also covers AI governance, risk management, compliance frameworks, red teaming, and operational practices needed for organizations deploying AI at scale. With practical examples and structured security approaches, the book helps cybersecurity professionals, engineers, and technology leaders build trustworthy, resilient, and responsible AI systems. 


BOOK]]>
AI Under Attack 
BY [Kris Kimmerle; David Okeyode]

Summary

AI Under Attack provides a practical guide to understanding and securing modern artificial intelligence systems against emerging cyber threats. The book explains why AI security is different from traditional cybersecurity, focusing on new risks created by large language models, generative AI, retrieval-augmented generation (RAG), and autonomous AI agents. 
It explores major attack techniques such as prompt injection, jailbreaks, data poisoning, memory and context manipulation, and exploitation of AI tools. The authors present methods for designing safer AI systems through threat modeling, secure architectures, defensive prompting, guardrails, and Zero Trust principles. The guide also covers AI governance, risk management, compliance frameworks, red teaming, and operational practices needed for organizations deploying AI at scale. With practical examples and structured security approaches, the book helps cybersecurity professionals, engineers, and technology leaders build trustworthy, resilient, and responsible AI systems. 


BOOK]]> https://mklab.gr/showthread.php?tid=1228 Tue, 21 Jul 2026 02:47:35 +0300 mklabgr]]> https://mklab.gr/showthread.php?tid=1228
Mastering Large Language Models 
BY [Ajay Rawat , Vardan Pathak]

Summary

The guide “Mastering Large Language Models: Architectures, Applications, and Real-World Deployments of Large Language Models” provides a practical introduction to understanding, building, and applying modern Large Language Models (LLMs). It explains the foundations of LLM architectures, generative AI, prompt engineering, Retrieval-Augmented Generation (RAG), AI agents, and intelligent chatbot systems. The book focuses on moving beyond theory by presenting hands-on projects and real-world deployment approaches for creating AI-powered applications. 

It explores how LLMs can be integrated into workflows, how agents can automate complex tasks, and how developers can design reliable AI systems. With examples, illustrations, and practical guidance, the guide is aimed at students, developers, and professionals who want to gain applied skills in modern AI technologies. It connects core machine learning concepts with current industry practices, showing how LLMs are transforming software development, business processes, and human–computer interaction.

BOOK]]>
Mastering Large Language Models 
BY [Ajay Rawat , Vardan Pathak]

Summary

The guide “Mastering Large Language Models: Architectures, Applications, and Real-World Deployments of Large Language Models” provides a practical introduction to understanding, building, and applying modern Large Language Models (LLMs). It explains the foundations of LLM architectures, generative AI, prompt engineering, Retrieval-Augmented Generation (RAG), AI agents, and intelligent chatbot systems. The book focuses on moving beyond theory by presenting hands-on projects and real-world deployment approaches for creating AI-powered applications. 

It explores how LLMs can be integrated into workflows, how agents can automate complex tasks, and how developers can design reliable AI systems. With examples, illustrations, and practical guidance, the guide is aimed at students, developers, and professionals who want to gain applied skills in modern AI technologies. It connects core machine learning concepts with current industry practices, showing how LLMs are transforming software development, business processes, and human–computer interaction.

BOOK]]> https://mklab.gr/showthread.php?tid=1227 Tue, 21 Jul 2026 02:39:12 +0300 mklabgr]]> https://mklab.gr/showthread.php?tid=1227 Sutskever's List: Foundational ideas of modern AI 
BY Richard Heimann

Summary

Sutskever’s List is a guided exploration of the foundational ideas and research breakthroughs that shaped modern artificial intelligence. Inspired by Ilya Sutskever’s influential collection of key AI papers, the book explains how major advances—from AlexNet and ResNet to sequence models, attention mechanisms, Transformers, and large-scale training—transformed deep learning from an experimental field into today’s powerful AI ecosystem. 

Rather than simply reviewing papers, it connects them into a broader story about scientific discovery, engineering decisions, scaling, and the evolution of AI thinking. The book also examines deeper questions about intelligence, reasoning, complexity, and AI safety, showing how technical progress is linked with philosophical and societal challenges. Written for readers with some AI background but not necessarily researchers, it provides clear explanations of difficult concepts while revealing the historical and human factors behind modern AI.


BOOK]]> Sutskever's List: Foundational ideas of modern AI 
BY Richard Heimann

Summary

Sutskever’s List is a guided exploration of the foundational ideas and research breakthroughs that shaped modern artificial intelligence. Inspired by Ilya Sutskever’s influential collection of key AI papers, the book explains how major advances—from AlexNet and ResNet to sequence models, attention mechanisms, Transformers, and large-scale training—transformed deep learning from an experimental field into today’s powerful AI ecosystem. 

Rather than simply reviewing papers, it connects them into a broader story about scientific discovery, engineering decisions, scaling, and the evolution of AI thinking. The book also examines deeper questions about intelligence, reasoning, complexity, and AI safety, showing how technical progress is linked with philosophical and societal challenges. Written for readers with some AI background but not necessarily researchers, it provides clear explanations of difficult concepts while revealing the historical and human factors behind modern AI.


BOOK]]> https://mklab.gr/showthread.php?tid=1226 Tue, 21 Jul 2026 02:34:27 +0300 mklabgr]]> https://mklab.gr/showthread.php?tid=1226 Atlas of AI 
BY [Kate Crawford]

Summary

Kate Crawford’s Atlas of AI challenges the common idea that artificial intelligence is a purely digital and neutral technology. Instead, she presents AI as a vast physical and political system built on the extraction of natural resources, human labor, and personal data. The book explores how AI depends on mining, energy consumption, global supply chains, and often invisible workers who label data and maintain automated systems.

 Crawford argues that algorithms are not objective machines but reflect human choices, social inequalities, and existing power structures. She examines issues such as bias in datasets, surveillance, environmental damage, and the concentration of power among large technology companies.

 Through a global investigation of AI’s hidden infrastructure, the author shows that the costs of AI extend far beyond software and algorithms. Ultimately, Atlas of AI calls for a more responsible and democratic approach to technology, where innovation is balanced with social justice, environmental sustainability, and accountability.

BOOK]]> Atlas of AI 
BY [Kate Crawford]

Summary

Kate Crawford’s Atlas of AI challenges the common idea that artificial intelligence is a purely digital and neutral technology. Instead, she presents AI as a vast physical and political system built on the extraction of natural resources, human labor, and personal data. The book explores how AI depends on mining, energy consumption, global supply chains, and often invisible workers who label data and maintain automated systems.

 Crawford argues that algorithms are not objective machines but reflect human choices, social inequalities, and existing power structures. She examines issues such as bias in datasets, surveillance, environmental damage, and the concentration of power among large technology companies.

 Through a global investigation of AI’s hidden infrastructure, the author shows that the costs of AI extend far beyond software and algorithms. Ultimately, Atlas of AI calls for a more responsible and democratic approach to technology, where innovation is balanced with social justice, environmental sustainability, and accountability.

BOOK]]> https://mklab.gr/showthread.php?tid=922 Tue, 07 Jul 2026 17:02:14 +0300 mklabgr]]> https://mklab.gr/showthread.php?tid=922 Machine Learning Contests: A Guidebook 
by Wang He


Summary

Machine Learning Contests: A Guidebook, authored by Wang He, Peng Liu, and Qian Qian, serves as a comprehensive roadmap for navigating competitive data science and algorithmic challenges. Written by elite "competition professionals" with extensive championship experience, the text bridges the gap between theoretical knowledge and practical application. It systematically breaks down the entire lifecycle of a standard predictive modeling contest, taking readers through essential phases such as problem modeling, data exploration, rigorous feature engineering, and advanced model training. By translating chaotic real-world data into structured, actionable problem statements, the authors demystify the exact strategic pipelines used by top-tier competitors to achieve winning results.

Beyond foundational workflows, the guidebook delves into specific, complex competition domains, offering tailored strategies for text computing, temporal forecasting, recommendation systems, and advertising click-through rate prediction. Rather than focusing solely on basic algorithms, the text emphasizes practical routines, robust techniques, and hidden pitfalls that often determine the leaderboard rankings. By combining deep learning principles with hands-on domain expertise, this work transforms mathematical theory into an engineering discipline. Mastering these competitive frameworks ultimately enables data scientists and machine learning engineers to build highly optimized, resilient models capable of solving complex, high-stakes industrial problems.

BOOK]]> Machine Learning Contests: A Guidebook 
by Wang He


Summary

Machine Learning Contests: A Guidebook, authored by Wang He, Peng Liu, and Qian Qian, serves as a comprehensive roadmap for navigating competitive data science and algorithmic challenges. Written by elite "competition professionals" with extensive championship experience, the text bridges the gap between theoretical knowledge and practical application. It systematically breaks down the entire lifecycle of a standard predictive modeling contest, taking readers through essential phases such as problem modeling, data exploration, rigorous feature engineering, and advanced model training. By translating chaotic real-world data into structured, actionable problem statements, the authors demystify the exact strategic pipelines used by top-tier competitors to achieve winning results.

Beyond foundational workflows, the guidebook delves into specific, complex competition domains, offering tailored strategies for text computing, temporal forecasting, recommendation systems, and advertising click-through rate prediction. Rather than focusing solely on basic algorithms, the text emphasizes practical routines, robust techniques, and hidden pitfalls that often determine the leaderboard rankings. By combining deep learning principles with hands-on domain expertise, this work transforms mathematical theory into an engineering discipline. Mastering these competitive frameworks ultimately enables data scientists and machine learning engineers to build highly optimized, resilient models capable of solving complex, high-stakes industrial problems.

BOOK]]>