Aadam

@aadam.dev

Ph.D. student @ IUI. Studying #MachineLearning, #NeuroSymbolicAI

Really enjoying the seamless self-hosted sync for the Logseq DB version (set up with the help of Codex). Logseq nowadays acts as both my daily journal, and a place for LLM to write logs and context to. I'm loving the experience so far. Thank you Logseq team for developing and open-sourcing sync.

London #neuroscience people you may like this. We're hosting a series of talks at Imperial & Crick on how to get experiment and theory working together better. Each session will have a talk around this and extended networking / group discussion on the questions raised. Plus, free food! 🤖🧠🧪

Marcus Ghosh@marcusghosh.bsky.social · 6mo ago

How can we build better experiment-theory collaborations in neuroscience? We're hosting a series of 6 in-person, interactive workshops to discuss this. Come along! www.eventbrite.co.uk/e/understand...

👋Hello World Welcome to my humble digital abode! In my first post on this newly minted Ghost blog, I aim to articulate my goals and motivations behind this blog, and make some resolutions for myself and some promises to you (my dear reader or future self). Before delving into my motivations […]

Reclaiming my mind in the age of AI

👋 Hello World Welcome to my humble digital abode! In my first post on this newly minted Ghost blog, I aim to articulate my goals and motivations behind this blog, and make some resolutions for myself and some promises to you (my dear reader or future self). Before delving into my motivations, goals, and promises, let me first quickly introduce myself. My name is **Aadam** (yeah, that's my full name), and I'm currently doing a PhD in Computer Science. I won't say I'm quite an interesting fellow, but I do have some interests: 🤫 You can click on the highlighted text above to expand it further. There are some redeeming qualities to my character as well, but I'm not going to reveal all the goodies in our first meet-and-greet now, would I? You'll have to stick around (if intrigued) to find out more. # Why this Blog? To answer this question, I'll have to take you on a tangent and tell you a woeful story. Once upon a time there lived a starry-eyed boy, eager to learn and make his mark on the world. Inspired by sci-fi movies, and novels like I-Robot, he dreamed of creating truly intelligent machines one day. To realize his dream, he learned to code, enjoyed going through dense C++ manuals, and participated in several coding competitions representing his institution. Though with the passage of time, his interests shifted, his duties increased, and his priorities changed. Life happened. And more importantly, the world evolved. Suddenly, skills that were sought after and valued before were becoming obsolete. Even though AI wasn't truly intelligent yet, it became proficient enough to replace some of the skills that required intelligence before. Skills such as programming and writing, which required immense effort before, were being delegated to AI Agents. And more importantly, if you didn't use these new technologies, you'll get left behind. So, with time, he started relying on these technologies, and stopped developing and reinforcing his own skills in that particular domain. And slowly but surely, his skills atrophied. This is one of the main reasons behind the "Why" of this blog for me. I don't want my skills and my capabilities to fade away. I want to practice and improve my writing skills, in a carefree, safe, and personal environment, where I don't have to worry about meeting deadlines or quotas. I can polish my skills at my own pace, writing about what I want, and developing my own voice. There will surely be many errors (given that English isn't my first language), but that's fine. After all, you ~~only~~ mainly learn from your mistakes. I don't want to be entirely dependent on AI tools for writing and thinking. > “Once men turned their thinking over to machines in the hope that this would set them free. But that only permitted other men with machines to enslave them.” > > ― **Frank Herbert,**Dune I know that AI agents will be prevalent in the foreseeable future, and they are just tools (not actually intelligent beings for now 🤨), and we should use them to efficiently perform our tasks. They have their roles in writing, coding, brainstorming, re/searching, prototyping, and more, and can aid us in reaching our goals much more quickly and efficiently. I'm not against their usage. I regularly use them a lot to automate/skip mundane tasks. I just don't want to loose my own skills in the process. I have noticed this gradual skill decay personally and there have been some public reports on this as well. After the advent of calculator, it wasn't really necessary to memorize/practice complex calculations when you can simply get the answer quickly. That skill isn't required anymore. I wonder what skills will get obsolete after the Agentic AI era. So, the aim of this blog is quite selfish I'd say. I just want to develop my writing skills. I want to be able to confidently articulate my thoughts for a public audience. I really enjoyed the following quote by Brandon Sanderson (one of my favorite fantasy authors) in his recent talk where he discussed why he doesn't consider the AI generated output to be "Art". > Remember art is not just the story. It is not just the painting or the sculpture or whatever else you love to create. It's also the process of creation and what that process did to you. We make art because we can't help it. It's part of us. We understand what it is. We are drawn to it because we are of the same substance. We are the arts. > > Brandon Sanderson – We Are The Art | Brandon Sanderson’s Keynote Speech The basic idea is that "Art" isn't the end product (a generated poem, drawing, painting, novel), but the journey one took to get to that end product. And that's why I'm starting this blog. To go on a journey to rediscover myself and redevelop my skills. To share what I learn along the way. To revel in the joy of writing, living, and learning. # What to expect? 100% Human generated, error-prone prose. That's my only commitment to both myself and you, my dear reader. Again, if I use AI, then it would defeat the whole purpose of this blog. So, from brainstorming, to outlining, to writing, and finally editing, everything will be done by me and me alone. And this is a personal blog, so don't expect any adherence to a specific niche topic. I'll write about whatever topic catches my attention at that moment. I'm mainly interested in: Artificial Intelligence (Machine Learning, Deep Learning, Reinforcement Learning), Note-taking (Obsidian, Logseq, AnyType, Thymer, Tana), Programming (Julia, Go, Python), Fantasy Novels, Academic Life, and more. * * * So, if you want to get to know me more, learn about the journey I'm embarking on, and track my progress on this exciting path, stick around, introduce yourself in the comments, and follow along. If not, I still thank you for reading my incoherent thoughts and sticking till the end of this post. Looking forward to writing and sharing more, Insha'Allah.

aadam.dev

“Perception as Inference” is a century-old idea that has inspired all major theories in neuroscience 🧠, including: ✅ Sparse Coding ✅ Predictive Coding ✅ Free Energy Principle & more! In my new blog post, I build the intuition behind this idea from ground up 👉[1/6]🧵 🧠🤖🧠📈

Bild

Is anarchist science possible? As an experiment, we got together a large group of computational neuroscientists from around the world to work on a single project without top down direction. Read on to find out what happened. 🤖🧠🧪

Diagram of how the "collaborative modelling of the brain" (COMOB) project started. Starting material lead to group research or solo research, coming together in online workshops (monthly) in an iterative cycle, finishing with writing up together. The diagram is illustrated with colourful cartoon blob characters.

1/ What if we make robots that process touch the way our brains do? We found that Convolutional Recurrent Neural Networks (ConvRNNs) pass the NeuroAI Turing Test in currently available mouse somatosensory cortex data. New paper by @Yuchen @Nathan @anayebi.bsky.social and me!

Task-Optimized Convolutional Recurrent Networks Align with Tactile Processing in the Rodent Brain

Introducing Bases, a new core plugin that lets you turn any set of notes into a powerful database. Now available to everyone with Obsidian 1.9!

Bild

1/3) This may be a very important paper, it suggests that there are no prediction error encoding neurons in sensory areas of cortex: www.biorxiv.org/content/10.1... I personally am a big fan of the idea that cortical regions (allo and neo) are doing sequence prediction. But... 🧠📈 🧪

Sensory responses of visual cortical neurons are not prediction errors

Predictive coding is theorized to be a ubiquitous cortical process to explain sensory responses. It asserts that the brain continuously predicts sensory information and imposes those predictions on lo...

biorxiv.org

Hello world! This is the RL & Agents Reading Group We organise regular meetings to discuss recent papers in Reinforcement Learning (RL), Multi-Agent RL and related areas (open-ended learning, LLM agents, robotics, etc). Meetings take place online and are open to everyone 😊

Super excited to share our new paper! We spent the past years building an alternative AI approach, and now we demonstrate a whole range of advantages. Robust object & pose detection, generalization, data and compute efficient training, continual learning, shape bias, intelligent policies, and more!🦾

Thousand Brains Project@thousandbrains.org · last yr.

2/ The first paper: “Thousand-Brains Systems: Sensorimotor Intelligence for Rapid, Robust Learning & Inference”, is out now: arxiv.org/abs/2507.04494. It showcases our first working system, codename Monty.