├── LICENSE └── README.md /LICENSE: -------------------------------------------------------------------------------- 1 | MIT License 2 | 3 | Copyright (c) 2025 Mahbuba 4 | 5 | Permission is hereby granted, free of charge, to any person obtaining a copy 6 | of this software and associated documentation files (the "Software"), to deal 7 | in the Software without restriction, including without limitation the rights 8 | to use, copy, modify, merge, publish, distribute, sublicense, and/or sell 9 | copies of the Software, and to permit persons to whom the Software is 10 | furnished to do so, subject to the following conditions: 11 | 12 | The above copyright notice and this permission notice shall be included in all 13 | copies or substantial portions of the Software. 14 | 15 | THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 16 | IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 17 | FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE 18 | AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER 19 | LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, 20 | OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE 21 | SOFTWARE. 22 | -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- 1 | # Artificial Intelligence: Problem-Solving Search Strategies 2 | 3 | Have you ever wondered how Google Maps finds the fastest route for you? Or how a video game character knows where to go next? 4 | 5 | A key component of AI drives these capabilities. When AI needs to solve a problem, such as finding the best way to get from one place to another, it searches through different possibilities. This process is known as a Search Strategy in AI. 6 | 7 | Types of Search Strategies in AI 8 | 9 | AI search strategies are divided into two main types: Uninformed Search and Informed Search. 10 | 11 | ★ Uninformed Search: 12 | This involves exploring options systematically without prior knowledge or hints. 13 | 14 | For example, imagine you’re in a maze and systematically explore every path without knowing which one leads to the exit. 15 | 16 | Example: 17 | 1. Breadth-First Search (BFS): Explores all possible paths level by level. 18 | 19 | 2. Depth-First Search (DFS): Follows a single path all the way to the end before backtracking. 20 | 21 | Uninformed search methods are straightforward but can be time-consuming. 22 | 23 | ★ Informed Search: 24 | Here, AI applies strategies using some hints or knowledge about the best possible path. 25 | 26 | For example, Google Maps uses traffic data and distance estimates to calculate the shortest or fastest route. 27 | 28 | Example: 29 | 1. A* Search: Uses distance traveled and an estimated cost to the goal to find the most efficient route. 30 | 31 | Pathfinding in AI 32 | 33 | One of the most common applications of searching is "Pathfinding". 34 | 35 | Imagine a video game where your character needs to travel from the office to home. 36 | 37 | • If AI uses an uninformed method, it explores routes systematically without prioritizing the best option. 38 | • If AI uses an informed method, it leverages distance and time estimates to select the fastest route. 39 | 40 | Real-Life Applications 41 | 42 | Search strategies underpin everyday tools like navigation apps, online search engines and video game mechanics. 43 | 44 | Whenever you use a navigation app, play a video game or search for something online, AI's search algorithms are at work, making sure you get the best results efficiently! 45 | 46 | 47 | Reference Book: Forthcoming book titled "Artificial Intelligence: Echoes of a Silent Intellect", which will be published on Amazon soon. 48 | 49 | # Guidelines of Generative AI 50 | 51 | ★ Topics: In a structured order; from the most important to less critical topics! 52 | 53 | 1. GANs: DCGAN, StyleGAN, BigGAN. 54 | 55 | 2. Diffusion Models: DDPMs, Stable Diffusion. 56 | 57 | 2.1. Applications: Image Synthesis, Video Generation, Audio Generation. 58 | 59 | 3. Transformers: GPT, DALL-E, T5. 60 | 61 | 4. Text-to-Image Models: MidJourney, OpenAI's DALL-E. 62 | 63 | 5. VAEs: Applications in Image Generation & Compression. 64 | 65 | 6. Fine-Tuning & Custom Training: Domain-specific adaptations of pre-trained models. 66 | 67 | 7. Audio & Music Generation: WaveNet, Jukebox, Riffusion. 68 | 69 | 8. 3D & Video Generation: NeRF for 3D Modeling, GAN-based Video generation & editing Tools. 70 | 71 | 9. Ethics in Generative AI: Biases, Copyright, Safety Concerns. 72 | 73 | 10. Applications of Generative AI: Gaming, Content Creation, Drug Discovery, Digital Twins for Simulations. 74 | 75 | ★ Math for Gen AI 76 | 77 | 1. Linear Algebra: Matrices, Vectors, Eigenvalues, Eigenvectors. 78 | 79 | 2. Optimization: Gradient Descent, Convex Optimization, Backpropagation. 80 | 81 | 3. Probability & Statistics: Probability Distributions, Bayesian Inference, Hypothesis Testing. 82 | 83 | # Guidelines of Agentic AI 84 | Agentic AI mainly operates based on Four Foundational Pillars: Memory, Planning, Decision-Making and Autonomous Execution. 85 | 86 | ★ How to Get Started with Agentic AI? 87 | 88 | 1. Reinforcement Learning (RL) 89 | 90 | 2. Large Language Models (LLMs): Start with Hugging Face and OpenAI documentation. 91 | 92 | 3. Autonomous Systems & Multi-Agent AI: Start with the book "Multi-Agent Systems" and explore open-source Autonomous AI projects. 93 | 94 | 4. Memory & Planning Systems: Start with open-source projects like LangChain and AutoGPT. 95 | 96 | # Essential Topics of Agentic AI 97 | 98 | Most Important and Practical Aspects: 99 | 100 | 1. Core Algorithms in Agentic AI 101 | 102 | • RL: DQN, Policy Gradient Methods (PPO, A3C), Actor-Critic Models. 103 | 104 | • MAS: Collaboration and Competition between agents. 105 | 106 | 2. Applications of Agentic AI 107 | 108 | • Autonomous Vehicles: Self-driving cars and drones (Tesla, Waymo). 109 | 110 | • Robotics: Robots for industrial automation (e.g., Boston Dynamics). 111 | 112 | • Gaming AI: Adaptive NPC behavior (e.g., AlphaGo, AlphaStar). 113 | 114 | 3. Planning & Decision Making: MDPs, MCTS. 115 | 116 | 4. Human-Agent Interaction: Natural Language Communication (Alexa, Siri), Emotional Intelligence (but still emerging). 117 | 118 | 5. Ethical Concerns: Safety in autonomous systems, Bias in AI decision-making. 119 | 120 | 6. Tools to Start With 121 | 122 | • OpenAI Gym: RL simulation. 123 | 124 | • Unity ML-Agents: Game-based AI training. 125 | 126 | • PyBullet: Robotics simulation. 127 | 128 | 129 | ★ Math for Agentic AI 130 | 131 | • Linear Algebra: For state representation (matrices, vectors). 132 | 133 | • Probability: For Markov processes and decision-making. 134 | 135 | • Optimization: Gradient descent and policy optimization. 136 | 137 | 138 | # Quantum Computing 139 | 140 | Quantum computing is currently a hot topic in the tech world. 141 | 142 | Physicist Richard Feynman first proposed that by harnessing the principles of quantum mechanics, we could construct computers that operate exponentially faster than classical ones. In 1994, a mathematician named Peter Shor developed "Shor's Algorithm" which allows a quantum computer to factor large numbers very quickly, potentially breaking today’s encryption systems. Research in this field has been ongoing ever since. 143 | 144 | ★ What is Quantum Computing? How Does Quantum Work? 145 | 146 | The computers we use are classical computers. They operate using "bits" which are either 0 or 1. However, quantum computers work with "qubits". Qubits can hold multiple possibilities at once. For example, 50 qubits together can process over 2^50 possibilities (1.125 quadrillion). These qubits can be both 0 and 1 at the same time, a property called "Superposition." 147 | 148 | 1. Superposition: A qubit can be both 0 and 1 simultaneously. This allows a quantum computer to calculate many possibilities at once. 149 | 150 | 2. Entanglement: When two qubits are entangled or linked to each other, a change in one instantly changes the other, no matter the distance between them (this increases the computer’s speed even more). Einstein famously described this phenomenon as "Spooky Action at a Distance." 151 | 152 | 3. Interference: Interference uses quantum waves to identify the correct solution while eliminating incorrect possibilities. 153 | 154 | With these three principles, quantum computers can quickly solve complex problems that are impossible for classical computers. For instance, 100 qubits can process 2^100 possibilities which is more than the number of particles in the universe! 155 | 156 | ★ How Are Quantum Computers Built? What Challenges Are Faced? 157 | 158 | Building a quantum computer is extremely difficult and costs hundreds of millions of dollars. Qubits can be electrons, photons or superconducting circuits. They must operate at temperatures near absolute zero (-273°C). Moreover, qubits are highly sensitive; keeping them stable is tough and even slight movement, noise or heat can cause errors. That’s why they’re kept in special chambers. Additionally, quantum computing requires new types of algorithms which require time to develop and master. 159 | 160 | ★ Current Situation (Up to 2025): 161 | 162 | 1. Google announced "Quantum Supremacy" in 2019. Their Sycamore processor with 53 qubits, completed a task in 200 seconds that would take the world’s fastest supercomputer 10,000 years. By 2024, their Willow Chip has become even more advanced. 163 | 164 | 2. IBM’s Heron processor has 133 qubits and they’re providing researchers with cloud access through their "Quantum Center." Companies like IBM and Microsoft have brought quantum computing to the cloud, allowing ordinary people to try it, speeding up research. 165 | 166 | 3. China has set new records in quantum computing with their Jiuzhang 2.0, which uses photons. 167 | 168 | 169 | # Two Chrome Tools for Researchers 170 | 171 | 1. ExCITATION journal ranking in Google Scholar™️ 172 | 2. Rapid Journal Quality Check 173 | 174 | Two popular extension available on the Chrome Web Store that instantly show the quality and ranking of any academic journal. 175 | 176 | Once you install the extension, whenever you browse sites like Google Scholar, it automatically displays: 177 | 178 | Quartile Ranking (Q1 / Q2 / Q3 / Q4): You can immediately see which quartile the journal belongs to. 179 | 180 | How to Install? 181 | 182 | 1. Go to the Chrome Web Store 183 | 2. Search "Extension Name" 184 | 3. Click Add to Chrome 185 | 186 | After installation, the journal metrics will appear automatically on Google Scholar and other research platforms. 187 | 188 | 189 | # How to find a journal for your paper? 190 | 191 | Go to https://journalfinder.elsevier.com/ 192 | 193 | Write the abstract for your paper in it and search, you will get a list of related journals. 194 | 195 | If you submit your paper to a subscription-based journal (not open access), many of them do not charge any publication fee, because the cost is covered by institutional subscriptions. 196 | 197 | However, some journals still have submission fees or optional open-access charges, so it’s always important to check the journal’s "Author Guidelines" before submitting. 198 | 199 | 200 | 201 | 202 | 203 | 204 | --------------------------------------------------------------------------------