Browse the notes

Description: The Bandit-Based Approach to Hyperparameter Optimization is a method used in machine learning to efficiently and iteratively select the best hyperparameters for a mod…

Description Common Spatial Pattern (CSP) is a signal processing technique used in the field of EEG-based brain-computer interfaces (BCIs). It is a method to enhance the discrimina…

Description: Curriculum Learning is a machine-learning technique that mimics the way humans learn, starting with easier concepts and gradually progressing to more complex ones. It…

Description: Data Mining is a process of discovering patterns, trends, and useful information from large datasets. It involves various techniques to uncover hidden insights within…

Description EfficientNet is an advanced deep learning model known for its efficiency and effectiveness in image recognition tasks. Developed by Mingxing Tan and Quoc V. Le at Goog…

Description: Embedded Method for Feature Selection is a technique used in machine learning to select and prioritize relevant features during the model training process. This metho…

Feature engineering is the process of transforming raw data into features that are suitable for machine-learning models. In other words, it is the process of selecting, extracting…

Filter Method Description: Feature selection is a crucial step in machine learning to enhance model performance, reduce overfitting, and improve efficiency. The Filter Method is o…

Description This is a clustering algorithm that starts out with k classes and divides a set into k clusters. Applications K-means clustering algorithm has many applications, inclu…

Description Langchain for LLMs, or "Language Modeling Chain for Large Language Models," is a novel approach in the field of natural language processing (NLP) that leverages a seri…

Description: Learning Transferable Visual Models From Natural Language Supervision is a cutting-edge approach in machine learning and computer vision. This technique involves trai…

Description Linear Discriminant Analysis (LDA) is a dimensionality reduction technique used in machine learning and statistics. It's particularly useful for supervised classificat…

Overfitting and underfitting are the most common reasons why machine-learning algorithms perform poorly. Two metrics for this is the bias (the actual error rate), and the variance…

Description Proximal Policy Optimization (PPO) is a reinforcement learning algorithm introduced by OpenAI that aims to optimize policies in an efficient and stable manner. PPO bel…

Description: Reinforcement Learning (RL) is a type of machine learning paradigm where an agent learns to make sequences of decisions to accomplish a goal in an environment. The ag…

Description Riemannian geometry-based classifiers are a class of machine learning algorithms that leverage concepts from Riemannian geometry to perform classification tasks. These…

Description: Transfer Learning (TL) is a machine-learning technique where knowledge gained from solving one problem is applied to a different but related problem. In this method,…

Description The Wavelet Transform is a mathematical transform used for analyzing and processing signals and data. It decomposes a signal into different scales and frequencies, mak…

Description The Wrapper Method for feature selection is a technique used in the field of machine learning to select the most relevant features for a predictive model by evaluating…

Breadth first search is a graph-search algorithm that can be used for a variety of situations. Some examples include: - Finding the shortest route - Figuring out if a friend of a…

This is how compilation and execution work in C++: Start out with your C++ program, Then the preprocessor (the thing that works with the statements) pretty much adds in the C++ co…

In c++ streams are everywhere! This whole time, when you do: and do std::cout and std::cin you are reading/writing to streams! All of the operations with std::cout and std::cin yo…

The free store manager is the thing that allocates memory to the free store in c Something very important to know about some C++ keywords that regard the free store manager is tha…

Graphs are pervasive in computer science. Graphs can give a lot of information and be used to solve so many problems. Graphs can be defined by two sets, one with the vertices (the…

This is advice from this video on how to become great: 1. Learn how to debug code 2. Read code every day and try to understand it so you know how experienced programs code 3. Work…

Whenever you allocate memory to the free store you make an implicit contract with the free store manager that has three clauses: Clause 1: Having no pointer to a bit of allocated…

L-value is what is on the left side of an equals and it is a variables address, so you know where you are putting the contents of the expression into. See Also 1. computer-science…

But writing the lines to compile all of my source files takes so long 😩 Makefile has got you covered 😎 Basically, it is a way to "make" your project based on seeing which files…

Maps are a way of storing data with a key-value pair. C++: Use the map.at() function so that you don't accidentally create new key-value pairs in the map. The preferred way to pop…

These two different types of values explains a lot of what happens and why things are the way they are in C++ L-Values L-values are values that are on the left side of the equatio…

This is a topic in computer-science that whenever you acquire a resource that should happen in the initialization of the object, and whenever you release that resource that should…

Key Statement Sketching for datasets is a technique in computer science and data analysis that creates compact, memory-efficient representations (sketches) of large datasets, allo…

Switch statement compares one object to a bunch of other cases, these cases must be constant at compiler-time and look like: If the case does not have a break then it will continu…

TDD is where you use tests to make sure your code is operating correctly, you use and write these tests before or while you are writing the code. Making tests ensures your past co…

Type conversions is turning one thing into another, there are some automatic conversions that happen which are called implicit conversions and they happen when you do an operation…

All a type is in computer science is just a label that tells the compiler what kinds of operations, methods, and values that a specific amount of memory can do. An object is just…

A vector is pretty much like a list. The only thing I want to write about here is that you want to use: vector.at(index); instead of vector[index] because vector.at(index) will th…

Computer fatigue happens to me when I spend too long in front of my computer, especially sitting down. Here some things that can help: - Take breaks: Peppering your workday with s…

For a lot of things in life, the more time/effort/money etc you put in the less and less you get back, this is important to know when thinking about opportunity cost, since eventu…

Focus is an important part of learning because it makes everything else more efficient. Distractions are costly to the brain, making you work less efficiently, for shorter periods…

Focus System A system for sustained, distraction-free deep work. Built on the entry-point model, the flow cycle, and environment-first design. Core Philosophy The Entry-Point Mode…

This person finished top of his class while working significantly less. Aim High This is a very interesting way to do things. When you aim high it changes how you think and you st…

A beginner's guide to the keyboard setup people keep asking me about, and how to steal the parts you like. Written with AI: Claude Code drafted this guide from my actual config fi…

The thesis of this video is that you need to clear allostatic load consistently (the stress from constantly adjusting to life's pressures). It also suggests to change the way you…

This is advice from this video on how to become great: 1. Learn how to debug code 2. Read code every day and try to understand it so you know how experienced programs code 3. Work…

Main Steps: 🎯 Set Goals Define long-term, medium-term, and short-term goals to provide direction and purpose. 📋 Break Tasks Down Simplify goals into manageable weekly actions, a…

Matt's post-meeting automation system, and how to build your own This note was drafted by Claude Code on 2026-11-02 from the notes and code of the system it describes, which is wh…

Metalearning came to my attention from Scott Young's ultralearning and it is the process of learning about learning. In a general sense, like how I am learning about learning by r…

Perceived Importance Drives Flow The first benefit of work compression is clarity in the form of perceived importance. With less time, your brain starts automatically filtering fo…

From the pareto-principle would it be better to just set a time limit for studying. For example, I can only study for 2 hours for this test. This would challenge myself to find th…

Weekly Review System A 30-45 minute weekly operating system that produces clarity, relief, and actionable plans. Not a guilt-inducing audit. Related systems: ideal-week-system, ac…

The email app I recommend to anyone who lives in Gmail, and how to try it in 15 minutes. Written with AI: Claude Code drafted this guide from Superhuman's help pages and my own us…

"Goal is that my life is designed in a way so that I achieve what I achieve not through luck but by design." "A wise man does at once what a fool does finally" "don't listen to an…

Questions What is leptin and what does it do? Notes Insulin Insulin increases the number of glucose transporter in the cell membranes. This is GLUT-4 transporter. In the liver (GL…

Questions: What is every way that food preference and aversion develops? <detailsConditioned and uncondition stimulu & responses</details What is PKU (phenylketonuria)?<detailsis…

Questions What happens to an animal if the food it eats just gets drained away? <detailsNon-stop eating</details How does your GI track influence your feeling of satiety? Where do…

Module 1 psyc417-olfaction-taste psyc417-gustation Module 2 psyc417-gustation psyc-417-food-preference-aversion-learning psyc417-hunger psyc-417-gastrointestinal-signaling Module…

Questions Notes The goal of this lecture is to teach about what leads to obesity and how people tend to lose weight Drastic weight change is difficult. When you go on a diet to lo…

Questions Notes This is the conclusion of the course 😢, here are the takeways: - The brain and body is a system and systems interact. - You should know about the best conditions…

Questions Notes CS+ potentiated eating is seeing if an animal will learn to eat food when there is a cue and it is not hungry. In this, the CS+ induces a complex emotional or moti…

Questions while studying What is the whole pathway in regards to taste tongue to what??<details First order neural projects go to the NST (rostral nuclues of the solitary tract),…

Questions What is the hunger signal? <details </details What is lipoprivic eating? <detailsLipo-fat, privic- deprived, so eating when we are on a high fat diet</details What is gl…

Questions What is the arcuate nucleus? Notes Craniopharyngamia is a tumor between the pituitary gland and the hypothalamus. Pituitary adenoma is a benign tumor that can cause hype…

Questions while studying Olfaction What is the field of psychophysics? What are cranial nerves and what type of data do they transmit? What do G-protein coupled receptors have to…

Questions Notes Brain development is as follows: Neural Tube formation at wee 3, Basic Brain Region Formation Begins at week 10, Neuronal Migration Begins at Week 13, Structural F…

Questions Notes Natural rewards are rewards that are produced by sensory stimuli (fresh-air, cat). Non-natural reward are rewardsa produced by drug or brain stimulation. Rewards a…

Questions Notes Drinking too much water at once can cause water intoxication, drinking too much alcohol can just cause you to die. There are physiological and environmental things…

Notes on Mark Manson's "The Three Levels of Self-Awareness" Reading notes on Mark Manson's essay Why You Suck at Self-Awareness. Summary in my own bullets — the framework and exam…

Human Common Fundamental Needs Subsistence & Security Physical Needs - Air - Food - Health - Movement - Physical Safety - Rest / sleep - Shelter - Touch - Water - Sex Security - C…

Charisma is power, warmth and presence. Pause for 2 seconds before speaking. Have a downwards intonation at end of speech. Don't nod that much. Exercises: Try to become more prese…

The cobra effect happens when a solution to a problem ends up making the problem worse or introducing another, worse problem[^1] See Also 1. Behavior 2. Buisness 3. psychology [^1…

The Dunning Kruger effect happens when the knowledge to evaluate ones performance and the knowledge required for performance is not substantial or the same. [^1] Incompetent peopl…

The false consensus is the normal human idea that other people are similar to us (i.e. have similar experiences/knowledge/etc) to us even when that is not true. It is a reason why…

The generation effect is that if you generate something you remember and learn it better. What was also cool (but I don't believe it that much) is that after learning about the ge…

Infotrophism is the act of trying to maximize information. The human brain does this is very cool ways. One way is to "subtract" past information to make the brain more sensitive…

The more often you have a specific thought pattern (positive thoughts vs negative thoughts) the easier it will be to have that specific thought pattern. See Also 1. psychology 2.…

A memory I have about the nocebo effect is that even police officers fall victim to the nocebo effect, they believe that fentanyl will kill them, so if they believe there is fenta…

Perceived Importance Drives Flow The first benefit of work compression is clarity in the form of perceived importance. With less time, your brain starts automatically filtering fo…

The placebo effect is basically the power-of-belief and is used a lot in the medical field where patients are giving medicine that has no active ingredient. Because people have di…

rejection-proof Purpose of the rejection challenge: I want to improve my insecurities, which I think stem from my fear of rejection, being judged, people pleasing, etc. I think do…

Computer fatigue happens to me when I spend too long in front of my computer, especially sitting down. Here some things that can help: - Take breaks: Peppering your workday with s…

From a Youtube video I watched the guy talked a lot about how our bodies know what we need to some degree and crave that food. One reason for this is the amino-static hypothesis w…

Questions Notes Progressive Overload Increase either weights or reps (I'd increase weight). You could also make the exercise more difficult by controlling the negative. Amount of…

Notes Stats Gains do not grow linearly with time spent in the gym or volume. For health, you can maximize health benefits with 30-60 minutes of weight training per week! 1-4 Sets…

Here is going to be my collection of thoughts about longevity. Measurement is key If you want to predict how long you're going to live, maximize that number, or slow down aging as…

Questions Notes Psychological stress impacts performance by decreasing energy. Anything that helps remove this stress improves performance in the gym. See Also 1. https://www.yout…

General A good morning routine is something that is a Habits and optimizes your physical-health and mental-health health as well as setting myself up for success for the day. A go…

General A good night routine is a Habits that optimizes my physical-health and mental-health, involves reflection of the day, planning for the next day, and relaxation. My night r…

Running is SUPER good for mental-health and physical-health The heart rate zones are really important while running, and there are two main ways to train cardio Zone-2 Training (S…

Apparently standing is better for your physical-health compared to sitting down. It can help with computer fatigure and in general is bettere for your health. See Also 1.

Letter of Reference System A complete playbook for requesting, supporting, following up on, and maintaining letters of recommendation — built for someone applying to AI safety fel…

Cheap Flights System A comprehensive, research-backed system for booking flights at the lowest possible price. Built from aggregated data across Expedia Air Hacks (2025 & 2026), G…

Focus System A system for sustained, distraction-free deep work. Built on the entry-point model, the flow cycle, and environment-first design. Core Philosophy The Entry-Point Mode…

A beginner's guide to the keyboard setup people keep asking me about, and how to steal the parts you like. Written with AI: Claude Code drafted this guide from my actual config fi…

Learning System A system for turning articles, books, and content into durable knowledge that changes how I think and act. Built on spaced-repetition-science, deep-reading-and-enc…

Luck Operating System The luck-system describes what luck is and how it compounds. This file describes how Matt operates that system — the subsystems, projects, data models, caden…

Luck System Thesis: Luck is not random. The expected frequency of positive-sum unplanned outcomes — "lucky breaks" — is a function of durable choices about what you do, who can se…

Matt's post-meeting automation system, and how to build your own This note was drafted by Claude Code on 2026-11-02 from the notes and code of the system it describes, which is wh…

Speed As Default Thesis: Quick-to-respond is a generally dominant strategy. Ben Kuhn, "Impatient": faster replies get more feedback, more opportunities, and pleasantly surprise pe…

Weekly Review System A 30-45 minute weekly operating system that produces clarity, relief, and actionable plans. Not a guilt-inducing audit. Related systems: ideal-week-system, ac…

Description: Curriculum Learning is a machine-learning technique that mimics the way humans learn, starting with easier concepts and gradually progressing to more complex ones. It…

Description: Data Mining is a process of discovering patterns, trends, and useful information from large datasets. It involves various techniques to uncover hidden insights within…

Description EfficientNet is an advanced deep learning model known for its efficiency and effectiveness in image recognition tasks. Developed by Mingxing Tan and Quoc V. Le at Goog…

Description: Embedded Method for Feature Selection is a technique used in machine learning to select and prioritize relevant features during the model training process. This metho…

Description Kernel Support Vector Machine (Kernel SVM) is a machine learning algorithm used for classification and regression tasks. It's an extension of the traditional Support V…

Description Langchain for LLMs, or "Language Modeling Chain for Large Language Models," is a novel approach in the field of natural language processing (NLP) that leverages a seri…

Description: Learning Transferable Visual Models From Natural Language Supervision is a cutting-edge approach in machine learning and computer vision. This technique involves trai…

Description: Transfer Learning (TL) is a machine-learning technique where knowledge gained from solving one problem is applied to a different but related problem. In this method,…

Summary This is a book that is all about one method for learning how to learn as fast as possible as well as giving advice for how to learn quickly and efficiently. It gives a ste…

The same as a flow integral, but when the start point (x,y) is the same as the end point (x,y). See Also 1. Multivariable calculus 2. flow-integral

A conservative vector field is any function f, where f can be written as the gradient of another function. It has a special property that the integral from points a to b of the cu…

Divergence Theorem is similar to greens-theorem but one dimension higher. See Also 1. Multivariable calculus

Flow integral is how much the vector field goes along the curve. It is intimately related to curl. It is the same as the line integral around the curve See Also 1. Multivariable c…

Flux in a physical sense is how much something is going through somearea. The more that something goes through somearea the higher the flow or the more that somearea increases the…

Flux is kind of the opposite of flow. It is intimately related to divergence and instead of the integral in the same direction of a curve, it is the integral in the Normal Directi…

Green's Theorem is the idea of relating what happens on the boundary of a curve to what happens inside of the curve❗ Proof Imagine a curve C that looks like: Well, you can actuall…

The line integral can integrate over Scalar Fields and Vector Fields. For a scalar field it can be thought of just normal integration of a function of a line, but that line has be…

The stokes' theorem is just an upgraded version of greens-theorem but for multiple dimensions. You can think that greens-theorem is Stokes' Theorem when working on a surface livin…

Highlight a claim, ask for a source, click Allow, and the link shows up on your words. Written with AI: Claude Code drafted this guide from a recorded run of the setup on my own G…

Backpropogation is the way that deep learning neural networks learn. The whole idea behind it is to try to figure out how to nudge each layer so that the cost function is minimize…

Description Component-based Analysis of Language Models (CALM) involves dissecting language models to understand how different components contribute to the overall performance and…

Matt's post-meeting automation system, and how to build your own This note was drafted by Claude Code on 2026-11-02 from the notes and code of the system it describes, which is wh…

Description: Reinforcement Learning (RL) is a type of machine learning paradigm where an agent learns to make sequences of decisions to accomplish a goal in an environment. The ag…

Description: Transfer Learning (TL) is a machine-learning technique where knowledge gained from solving one problem is applied to a different but related problem. In this method,…

AI isn't always the solution There are solutions already for many many problems, throwing AI adds many more problems than solutions. See Also 1. ai 2. sigaida

The cobra effect happens when a solution to a problem ends up making the problem worse or introducing another, worse problem[^1] See Also 1. Behavior 2. Buisness 3. psychology [^1…

The Dunning Kruger effect happens when the knowledge to evaluate ones performance and the knowledge required for performance is not substantial or the same. [^1] Incompetent peopl…

The false consensus is the normal human idea that other people are similar to us (i.e. have similar experiences/knowledge/etc) to us even when that is not true. It is a reason why…

A memory I have about the nocebo effect is that even police officers fall victim to the nocebo effect, they believe that fentanyl will kill them, so if they believe there is fenta…

The placebo effect is basically the power-of-belief and is used a lot in the medical field where patients are giving medicine that has no active ingredient. Because people have di…

Regression to the mean is a statistical occurrence that happens in day-to-day life. It is the reason why people attribute casual relationships to things that are just correlated.…

Notes on Mark Manson's "The Three Levels of Self-Awareness" Reading notes on Mark Manson's essay Why You Suck at Self-Awareness. Summary in my own bullets — the framework and exam…

What are some things that my current self can do that my future self would be greatful for? Scratchpad Follow up, what are things that compound over time, (things that start small…

Life Audit How To Identifying Core Values Start by listing your top 5-10 core values. These values are the base of your life vision. Think about things like making meaningful chan…

If there exists an action that I am afraid to do, but I know is good for me to do, others advise me to do, then I must do it moreso because I am afraid. This is for several reason…

Memento Mori - Write your own eulogy. Many people have heard of this, but I recommend writing two versions: Write the eulogy for if you died today, and then write the one for if y…

rejection-proof Purpose of the rejection challenge: I want to improve my insecurities, which I think stem from my fear of rejection, being judged, people pleasing, etc. I think do…

Computer fatigue happens to me when I spend too long in front of my computer, especially sitting down. Here some things that can help: - Take breaks: Peppering your workday with s…

The thesis of this video is that you need to clear allostatic load consistently (the stress from constantly adjusting to life's pressures). It also suggests to change the way you…

General A good morning routine is something that is a Habits and optimizes your physical-health and mental-health health as well as setting myself up for success for the day. A go…

General A good night routine is a Habits that optimizes my physical-health and mental-health, involves reflection of the day, planning for the next day, and relaxation. My night r…

Running is SUPER good for mental-health and physical-health The heart rate zones are really important while running, and there are two main ways to train cardio Zone-2 Training (S…

The cobra effect happens when a solution to a problem ends up making the problem worse or introducing another, worse problem[^1] See Also 1. Behavior 2. Buisness 3. psychology [^1…

For a lot of things in life, the more time/effort/money etc you put in the less and less you get back, this is important to know when thinking about opportunity cost, since eventu…

meta-skill decision-making https://www.youtube.com/watch?v=S5SBcAuqCSg&t=1s 5 Mental Models to Think like a Strategic Genius What is a mental model? It is simply a way of thinking…

An important concept to know about from economics, whenever you spend a finite resource, there is the cost of that resource, but there is also a cost that is the sum of everything…

Regression to the mean is a statistical occurrence that happens in day-to-day life. It is the reason why people attribute casual relationships to things that are just correlated.…

Description Deep Metric Learning is a subfield of machine learning that focuses on learning similarity metrics from data. It's particularly useful when you want to measure the sim…

Overfitting and underfitting are the most common reasons why machine-learning algorithms perform poorly. Two metrics for this is the bias (the actual error rate), and the variance…

Description: Transfer Learning (TL) is a machine-learning technique where knowledge gained from solving one problem is applied to a different but related problem. In this method,…

Description: In machine learning, the quality and characteristics of the dataset play a critical role in the performance of AI models. Here are some attributes of a dataset that c…

Computer fatigue happens to me when I spend too long in front of my computer, especially sitting down. Here some things that can help: - Take breaks: Peppering your workday with s…

General A good morning routine is something that is a Habits and optimizes your physical-health and mental-health health as well as setting myself up for success for the day. A go…

Resistance training is SUPER important for asthetics, becoming stronger, as well as preventing injury in live. See Also 1. exercise

Studying is the process of getting as much information about subject to stay in your mind as possible and having it stick. There are ways to improve studying such as: Sleep Sleep…

Chunking in the context of problem-solving is where you break problems down into smaller and smaller pieces until they are actionable steps. Chunking in the context of learning is…

Directness is the best way to learn generally, it turn theory to reality. The technique is to try to make the learning environment as close to the testing environment as possible[…

The generation effect is that if you generate something you remember and learn it better. What was also cool (but I don't believe it that much) is that after learning about the ge…

Learning System A system for turning articles, books, and content into durable knowledge that changes how I think and act. Built on spaced-repetition-science, deep-reading-and-enc…

Description Component-based Analysis of Language Models (CALM) involves dissecting language models to understand how different components contribute to the overall performance and…

Description: Curriculum Learning is a machine-learning technique that mimics the way humans learn, starting with easier concepts and gradually progressing to more complex ones. It…

Description Langchain for LLMs, or "Language Modeling Chain for Large Language Models," is a novel approach in the field of natural language processing (NLP) that leverages a seri…

Description: Learning Transferable Visual Models From Natural Language Supervision is a cutting-edge approach in machine learning and computer vision. This technique involves trai…

Human Common Fundamental Needs Subsistence & Security Physical Needs - Air - Food - Health - Movement - Physical Safety - Rest / sleep - Shelter - Touch - Water - Sex Security - C…

If there exists an action that I am afraid to do, but I know is good for me to do, others advise me to do, then I must do it moreso because I am afraid. This is for several reason…

Make Your Own Luck Serendipity is something you can manufacture. For creative serendipity, you take an unexpected event, look more into it, and see if there's an opportunity there…

My Answers to the Most Important Questions What do you think are the most pressing problems in the world? I'm defining pressing as the problems that are the combination of the mos…

Description: A Kalman Filter (KF) is a recursive, efficient, and versatile algorithm used for estimating the state of a linear dynamic system in the presence of noise and uncertai…

Description: Reinforcement Learning (RL) is a type of machine learning paradigm where an agent learns to make sequences of decisions to accomplish a goal in an environment. The ag…

Key Statement Sketching for datasets is a technique in computer science and data analysis that creates compact, memory-efficient representations (sketches) of large datasets, allo…

Letter of Reference System A complete playbook for requesting, supporting, following up on, and maintaining letters of recommendation — built for someone applying to AI safety fel…

Luck Operating System The luck-system describes what luck is and how it compounds. This file describes how Matt operates that system — the subsystems, projects, data models, caden…

Luck System Thesis: Luck is not random. The expected frequency of positive-sum unplanned outcomes — "lucky breaks" — is a function of durable choices about what you do, who can se…

Letter of Reference System A complete playbook for requesting, supporting, following up on, and maintaining letters of recommendation — built for someone applying to AI safety fel…

Charisma is power, warmth and presence. Pause for 2 seconds before speaking. Have a downwards intonation at end of speech. Don't nod that much. Exercises: Try to become more prese…

Framework for Asking A graceful request balances clarity (so they know exactly what you need) with autonomy (so they don't feel pressured). When you give people a clear "exit" or…

Description: Curriculum Learning is a machine-learning technique that mimics the way humans learn, starting with easier concepts and gradually progressing to more complex ones. It…

Description Deep Metric Learning is a subfield of machine learning that focuses on learning similarity metrics from data. It's particularly useful when you want to measure the sim…

Description: Learning Transferable Visual Models From Natural Language Supervision is a cutting-edge approach in machine learning and computer vision. This technique involves trai…

Description: The Bandit-Based Approach to Hyperparameter Optimization is a method used in machine learning to efficiently and iteratively select the best hyperparameters for a mod…

Description: Embedded Method for Feature Selection is a technique used in machine learning to select and prioritize relevant features during the model training process. This metho…

Description The Wrapper Method for feature selection is a technique used in the field of machine learning to select the most relevant features for a predictive model by evaluating…

Description Kernel Support Vector Machine (Kernel SVM) is a machine learning algorithm used for classification and regression tasks. It's an extension of the traditional Support V…

Model-Based Reinforcement Learning Description: Model-Based Reinforcement Learning (MBRL) is a subfield of reinforcement learning that combines model learning with control. In MBR…

Description Proximal Policy Optimization (PPO) is a reinforcement learning algorithm introduced by OpenAI that aims to optimize policies in an efficient and stable manner. PPO bel…

Description Linear Discriminant Analysis (LDA) is a dimensionality reduction technique used in machine learning and statistics. It's particularly useful for supervised classificat…

Key Statement Sketching for datasets is a technique in computer science and data analysis that creates compact, memory-efficient representations (sketches) of large datasets, allo…

Description The Wrapper Method for feature selection is a technique used in the field of machine learning to select the most relevant features for a predictive model by evaluating…

meta-skill decision-making https://www.youtube.com/watch?v=S5SBcAuqCSg&t=1s 5 Mental Models to Think like a Strategic Genius What is a mental model? It is simply a way of thinking…

Studying is the process of getting as much information about subject to stay in your mind as possible and having it stick. There are ways to improve studying such as: Sleep Sleep…

"Goal is that my life is designed in a way so that I achieve what I achieve not through luck but by design." "A wise man does at once what a fool does finally" "don't listen to an…

Description: Embedded Method for Feature Selection is a technique used in machine learning to select and prioritize relevant features during the model training process. This metho…

Filter Method Description: Feature selection is a crucial step in machine learning to enhance model performance, reduce overfitting, and improve efficiency. The Filter Method is o…

Description The Wrapper Method for feature selection is a technique used in the field of machine learning to select the most relevant features for a predictive model by evaluating…

Here is going to be my collection of thoughts about longevity. Measurement is key If you want to predict how long you're going to live, maximize that number, or slow down aging as…

Questions while studying Olfaction What is the field of psychophysics? What are cranial nerves and what type of data do they transmit? What do G-protein coupled receptors have to…

Apparently standing is better for your physical-health compared to sitting down. It can help with computer fatigure and in general is bettere for your health. See Also 1.

What are some things that my current self can do that my future self would be greatful for? Scratchpad Follow up, what are things that compound over time, (things that start small…

Life Audit How To Identifying Core Values Start by listing your top 5-10 core values. These values are the base of your life vision. Think about things like making meaningful chan…

Weekly Review System A 30-45 minute weekly operating system that produces clarity, relief, and actionable plans. Not a guilt-inducing audit. Related systems: ideal-week-system, ac…

Luck Operating System The luck-system describes what luck is and how it compounds. This file describes how Matt operates that system — the subsystems, projects, data models, caden…

Luck System Thesis: Luck is not random. The expected frequency of positive-sum unplanned outcomes — "lucky breaks" — is a function of durable choices about what you do, who can se…

Make Your Own Luck Serendipity is something you can manufacture. For creative serendipity, you take an unexpected event, look more into it, and see if there's an opportunity there…

This book deals with how the brain rewires itself to adapt to the environment. It introduces a lot of concepts such as information maximization (infotrophism) and the ideas of sen…

What is the point in learning something if you are going to forget it before it is useful? This what retention is all about. The Ebbinghaus's Forgetting Curve teaches us that memo…

Summary This is a book that is all about one method for learning how to learn as fast as possible as well as giving advice for how to learn quickly and efficiently. It gives a ste…

Luck Operating System The luck-system describes what luck is and how it compounds. This file describes how Matt operates that system — the subsystems, projects, data models, caden…

Luck System Thesis: Luck is not random. The expected frequency of positive-sum unplanned outcomes — "lucky breaks" — is a function of durable choices about what you do, who can se…

Make Your Own Luck Serendipity is something you can manufacture. For creative serendipity, you take an unexpected event, look more into it, and see if there's an opportunity there…

Description This paper presents a novel approach to EEG (Electroencephalography) signal analysis using a Convolutional Recurrent Attention Model (CRAM). The method focuses on subj…

Description Component-based Analysis of Language Models (CALM) involves dissecting language models to understand how different components contribute to the overall performance and…

Description: Transfer Learning (TL) is a machine-learning technique where knowledge gained from solving one problem is applied to a different but related problem. In this method,…

Computer fatigue happens to me when I spend too long in front of my computer, especially sitting down. Here some things that can help: - Take breaks: Peppering your workday with s…

From a Youtube video I watched the guy talked a lot about how our bodies know what we need to some degree and crave that food. One reason for this is the amino-static hypothesis w…

Questions while studying Olfaction What is the field of psychophysics? What are cranial nerves and what type of data do they transmit? What do G-protein coupled receptors have to…

Memento Mori - Write your own eulogy. Many people have heard of this, but I recommend writing two versions: Write the eulogy for if you died today, and then write the one for if y…

My Answers to the Most Important Questions What do you think are the most pressing problems in the world? I'm defining pressing as the problems that are the combination of the mos…

This is an interesting idea where people are secular (atheist) but also spiritual. It has an assumption that people "evolved" with music/religion/prayer/etc and they try to take a…

The more often you have a specific thought pattern (positive thoughts vs negative thoughts) the easier it will be to have that specific thought pattern. See Also 1. psychology 2.…

A memory I have about the nocebo effect is that even police officers fall victim to the nocebo effect, they believe that fentanyl will kill them, so if they believe there is fenta…

The placebo effect is basically the power-of-belief and is used a lot in the medical field where patients are giving medicine that has no active ingredient. Because people have di…

Description Common Spatial Pattern (CSP) is a signal processing technique used in the field of EEG-based brain-computer interfaces (BCIs). It is a method to enhance the discrimina…

Description Kernel Support Vector Machine (Kernel SVM) is a machine learning algorithm used for classification and regression tasks. It's an extension of the traditional Support V…

Description Linear Discriminant Analysis (LDA) is a dimensionality reduction technique used in machine learning and statistics. It's particularly useful for supervised classificat…

Before trying to solve a problem, actually try When something matters and you've been stuck for a while, the failure mode is rarely "I tried hard and it didn't work." It's that yo…

Chunking in the context of problem-solving is where you break problems down into smaller and smaller pieces until they are actionable steps. Chunking in the context of learning is…

meta-skill decision-making https://www.youtube.com/watch?v=S5SBcAuqCSg&t=1s 5 Mental Models to Think like a Strategic Genius What is a mental model? It is simply a way of thinking…

Description Common Spatial Pattern (CSP) is a signal processing technique used in the field of EEG-based brain-computer interfaces (BCIs). It is a method to enhance the discrimina…

Description: Power Spectral Density (PSD) measures the power distribution of a signal over its frequency components. It represents how the power of a signal is distributed over va…

Description The Wavelet Transform is a mathematical transform used for analyzing and processing signals and data. It decomposes a signal into different scales and frequencies, mak…

1-on-1 Norms I Endorse Here are 10 minutes of my thinking on norms I'd like to have for 1-on-1s I have: - Small talk: I endorse having a couple minutes of small talk / introductio…

Content I envision a system that, if I could hire someone to review my body camera footage alongside my calendar and typed communications – essentially a life log – could parse th…

2026 Goals My goals by the end of 2026 are: - Identify and fully commit to a long term career path - Create a system for self-improvement that allows me to double every 6 months.

About This Site I'm Matt, starting my career in AI Safety in the Bay Area, and very happy to welcome you to my personal website :) What will you find here? I like to keep organize…

7) Monthly Category Breakdown - 2025-09: - Shopping: 56.02% - Education: 16.60% - Housing: 15.55% - Health: 3.72% - Other: 2.78% - Software/Tools: 1.69% - Travel: 1.58% - Food: 0.…

Questions Notes Autoencoders are types of neural networks that comrpess the data into a latent space that can be used to reconstruct the previous data. Types Denoising autoencoder…

Black boxing is a way to learn things when you can use something without knowing how it works (its a black box). It is useful for competitive programming because you can learn how…

Bucket List This is a document that contains my "bucket-list", a list of everything I want out of life. This list may be old, outdated, not represent what I want right now, but I…

This is a breathing technique that is used to manage Stress and asthma. It has been studied and endorsed The Technique: Getting ready - A person should sit on the floor or on a ch…

Communities I Am a Part of The purpose of this document is to list out the communities I identify with, participate in, or am interested in. Personal Affiliation Scale 1- I am bar…

When copying an object in C++ it is sometimes useful to define a copy function instead of using the default, a default copy looks like: This results in a shallow copy where are al…

The Chebyshev's inequality states that at most N/k^2 \ (std({x}))^2 percent of the data can be outside k units of stdev away from the mean. To find the correlation coefficient you…

Powers of Languages L^i = L \ L^{i-1} Max elements of languages ^ i is len(L) ^ i Kleene star = $L^{} = L^{i},\forall i\in \mathbb{N}$ Important is that you can go from finite to…

NFA - DFA conversion. NFA can be converted to DFA because there are a finite number of states/edges in NFA. Given NFA M = (Q, \Sigma, s, \delta, A), create DFA M' = (Q^', ..) wher…

Current Supplements Active Stack | Supplement | Dose | Form | Timing | | -------------------- | --------- | ------------------ | --------------------------------------------------…

Date Me This is a document about dating me. WIP. Bio/Info Age: 21 Height: legally 6'0" (actually 5'11.5") Languages: English (native), Polish (conversational), French (conversatio…

I want to put in place a good eating routine because when you eat is also important, here is what I found: - Breakfast: Eating breakfast between 6:00 to 9:45 a.m. is recommended.…

Example of Using Cruxes for Ai Safety Uplift Organization here is an example exercise of me doing cruxes. you can see below a statement that i want to investigate if is true and w…

So, this is a common thread I've seen that someone being knowledgeable in a field does not make them good at sharing that knowledge. This is pretty frustrating because these peopl…

I wanted to but did not get into AI @ UIUC. I tried very hard but did not get into the silicon valley trip. I tried and really thought that I would get into Stanford TreeHacks but…

Failures This is a document where I document some of my failures. I don't like the word failures, but this is a document that more-so contains things I wanted to achieve but did n…

Feedback is important for learning because it helps you figure out where you are wrong/need to improve. The more immediate and specific the feedback the better. There are also dif…

In this article https://www.scotthyoung.com/blog/2015/10/07/strangely-useful-career-advice/, scott young gives a good idea that if you want to be an author and be sucessful, you s…

In order to find the parameterization of 2 surfaces intersecting, when they are 2 planes intersecting it is very easy, you just need to take the cross product of the two surfaces'…

Generated from Firefly III by scripts/finance/firefly-category-breakdown.py. Percentages only, by design — no absolute amounts are ever written here. 7) Monthly Category Breakdown…

Focusing The Focusing algorithm 1. Select something to bring into focus Something that’s alive for you, or that has been looming in the back of your mind Something that you haven’…

Friends Here are links to my friends: The purpose of this page is: - For you to see what my friends are doing - For you to make friends with them - I must believe their website/co…

Function Templates Function templates are SUPER useful, they are useful when you have a function that can be applied to multiple types of data, and you don't want to re-write the…

Gauss's law relates the electric flux through a surface and the charge within the surface, it is an application of the divergence-theorem. See Also 2. Physics

A distortion of the magnetic field that is created by objects that produce their own magnetic field. Much simpler to get rid of compared to soft-iron distortions where you just ne…

While watching this video the guy on the camera was talking about how your spongy bone tissue adapts to the forces that your body goes under in the day-to-day life. That reminded…

This is a video by a Youtuber that talks about things he wish he knew sooner. 1. Your 20's matter Your 20's is going to solidify the rest of your life. Things like your health hab…

common behavioral interveiw questions; make sure everything is unique; make sure you know what youre saying for the first minute; send them a thank you email UIUC groups for resea…

Intuition is just knowing something because it is just so obvious. This is built up when you understand how things work on a very deep level, using the Feynman Technique is one wa…

Questions Notes This is a learning algorithm that allows you to classify points. When there is an unlabeled point in the space the k-Nearest Neighbors take a "vote" to put the poi…

Learning follows diminishing returns, in order to mitigate this, you need to get new tools/a different learning experience or else you're going to waste time. This is why it is im…

Line Dances I Know Dances I Know | Name | Artist/Choreographer | Last Practiced | Tutorial/Demo | My Video | Notes | | -------------------------------------- | -------------------…

Symmetric A transpose = A Vector Space Defining by example, a vector space is a space that is defined by N vectors such that you can multiply and add vectors in that space and the…

Links anonymous feedback form schedule time with me website Twitter LinkedIn blog GitHub email

Questions Notes Matrix multiplication is used EVERYWHERE, there are 3 ways to multiply out a matrix. 1. Entry by entry. To multiply matrix A by B, take every row in A and multiply…

Take a seat and set a timer for 15 minutes. - Close your eyes and focus on your breath. - Count to five as you inhale and five as you exhale. - After a few minutes, relax from cou…

Most Important Life Skills This is a list of what I believe to be the most important life skills. These are not in any specific order. I just like to maintain this list and reflec…

The natural history of a disease is the natural progression of the disease in absence of treatment. It is in part the reason why Natural medicine works and people use it to scam p…

Title: The Natural Planning Method (Getting Things Done) The Natural Planning Method is the process the human mind instinctively follows to create and execute plans. Outlined in D…

A neuron doesn't know what it represents, this means it can represent any variety of spaces. This is what allows specialization. https://x.com/davideagleman/status/187084675787931…

Now London. I'm here this week for Liftoff Founders Week. I'm exploring founding a company in the AI safety space, and Liftoff is a program for that. I'm normally in Berkeley. Wha…

Office Hours Here are my office hours. If you want help with anything that you believe I can help you with, then please sign up! I want to help you. Before this meeting, please id…

Questions while studying What does it mean for a function to be onto? Each element of the co-domain corresponds to an element in the domain of the function. What does idA(x) mean?…

This principle just states that 80 percent of the X comes from 20 percent of the Y See Also 1.

In computer science, pass by reference is when you pass a reference of a variable to a function. Just like how you can pass a reference of a file to multiple people, when they edi…

In order to be a self-actualized and positive member of society, carle rodgers stated that we must live in an environment with the following qualities: openess, opportunities for…

Principles Growth is Most Important You cannot accomplish anything without growth. You are born as a baby with absolutely nothing. If you did not grow you would be crawling around…

You can use gprof and g++ to see how your code is running, use the -pg Then after running the code type: gprof testgprof gmon.out analysis.txt To look at the analysis See Also 1.…

This note will go over good programming practices, habits, thought patterns: Code is for people. Don't spend unnecessary time memorizing syntax and the like, learn how to google w…

These notes contain my best prompts: Teacher Your instruction is to act as a teacher and help me study for my [] . You will use your understanding of the topic as well as your abi…

Questions Why do hospitals only give medication and not other ways of treating. Notes This was such a good presentation! What a charismatic speaker! To get into a psychiatric hosp…

Questions 1. If narratives are mostly unconscious then would making us more conscious about that allow us to become better people? 2. Why 3. It seems like the unit for a person's…

Todo Questions Notes Tthe names for all of the parts of personality are Actor, Author, and Agent. Allport said that traits have a frequency, range of situations, and intensitty. C…

Personal Newsletter — How to Use and Write This folder is where the content of every newsletter issue lives. The CLI tooling lives at ~/Projects/personal-newsletter/ and reads fro…

Questions While Studying What is a relation? It just relates to elements in a set A. Formal definition is that when you apply a relation to a set then that relation R is the set o…

This is a sensory addition device that will tell the user which direction is north. It is using a Seeeduino XIAO, a 9-Axis IMU, and vibration motors to tell the user that north is…

There is reason to think that engineered minds could enjoy much greater durations and intensity of pleasure. Human psychology has evolved to generate pleasure and pain where this…

This is when the magnetic field is stretched or bent by metals nearby altering the shape of the magnetic field. A final calibration is usually needed in order to mitigate this eff…

Standards This is a document where I have put all of my standards for different areas of my life. These are the minimum acceptable levels of performance or behavior that I expect…

Questions Notes In linear algebra, you can have an equation: 3x + 5y + 0z = 7 0x + 3y + 4z = 5 -x + -2y + -1z = 1 And you can rewrite this as a matrix 3 5 0 7 0 3 4 = 5 -1 -2 -1 1…

Temp I had a problem in one of my RSO's where I couldn't find the solution to a part of my project. I had to implement BART to predict the relatedness of two sequences of letters.…

vLLM High-throughput, memory-efficient LLM serving with PagedAttention vLLM is an open-source inference and serving engine designed to maximize LLM throughput by eliminating KV-ca…

Vector Database A vector database is a type of database that is optimized for storing and querying high-dimensional vectors. These vectors are often used in machine learning and d…

Why You Can't Get Out of Bed in the Morning This is a oyutube video that talks about why you can't get out of bed in the morning, some tips it gives is to put your alarm out of re…