- Видео 440
- Просмотров 13 730
Orthogonal Research and Education Laboratory
Добавлен 6 июн 2016
Videos of talks and video content sponsored by Orthogonal Research (orthogonal-research.weebly.com).
Saturday Morning NeuroSim, 9-28
Attendees: Jesse Parent, Bradly Alicea, and Morgan Hough
Meta Connect, the Orion AR glasses, and its wrist-based neural interface. History of CTRL-Labs and NeuroTechX. Directory for Open Scholarly Communication (ASAPBio). HIBALL conference and the Big Brain project. Open theoretical questions in Complexity Science. What Cybernetics is and where it has gone, Part I. Averaging as Neuroscience Fiction. How to explore your data beyond the parametric. Self-replication and Neuromorphic completeness for bio-inspired computational theory.
References:
Kobak et.al (2016). Demixed principal component analysis of neural population data. eLife, doi:10.7554/eLife.10989.
Latimer et.al (2015). Single-trial ...
Meta Connect, the Orion AR glasses, and its wrist-based neural interface. History of CTRL-Labs and NeuroTechX. Directory for Open Scholarly Communication (ASAPBio). HIBALL conference and the Big Brain project. Open theoretical questions in Complexity Science. What Cybernetics is and where it has gone, Part I. Averaging as Neuroscience Fiction. How to explore your data beyond the parametric. Self-replication and Neuromorphic completeness for bio-inspired computational theory.
References:
Kobak et.al (2016). Demixed principal component analysis of neural population data. eLife, doi:10.7554/eLife.10989.
Latimer et.al (2015). Single-trial ...
Просмотров: 5
Видео
Cybernetics Reading Group, 9-26
Просмотров 269 часов назад
Attendees: Jesse Parent, Morgan Hough, Amanda Nelson, and Bradly Alicea. Reviewing selected references in the paper "Every Good Regulator Provides a World Model for Intelligent Systems". Hoel, "When the Map is Better than the Territory". Micro/Macro, Good regulations and non-homeomorphic maps, causality relations. Ha and Schmidthuber, "World Models". Markovian and Non-Markovian worlds, Reinforc...
Cognition Futures Reading Group, 9-25
Просмотров 1412 часов назад
Attendees: Jesse Parent, Amanda Nelson, Morgan Hough, and Bradly Alicea. Review of the paper "Behavior, Purpose, and Teleology" with reference to defining intelligent machines, the philosophy/history of complexity theory, and a hierarchy of behavioral teleology. The role of embodiment, enactivism, and predictive processing in Rosebluth et.al's behavioral teleology hierarchy. Rosenbluth et.al (1...
Saturday Morning NeuroSim, 9-21
Просмотров 821 час назад
Attendees: Bradly Alicea, Trevor Tanner, Shubham Soni, Morgan Hough, and Jesse Parent. Review of the #computational-psychiatry Slack channel. Genomics and TMS for Schizophrenia and Depression. Invisible illness. Gut microbiomics and mental health. Brain network metrics and multiscale neurodevelopment. Adolescent psychiatry. Reviewing the Table of Contents for "The Routledge Handbook of Philosop...
Cognition Futures Reading Group, 9-18
Просмотров 8День назад
Attendees: Jesse Parent and Bradly Alicea. Reviewing selected references in the paper "Every Good Regulator Provides a World Model for Intelligent Systems". W.R. Ashby references, the Homeostat, and the Ashby Box. The ethical regulator. Early Machine Learning concepts in the Cybernetics literature. PID control as Active Inference. Black Boxes and White Boxes for complex systems. Alicea et.al (2...
LLAMOSC (LLM Powered Agent-based Model for Open Source Community Sustainability) Demo
Просмотров 8514 дней назад
45-second demo of LLAMOSC (LLM Powered Agent-based Model for Open- Source Community Sustainability) Project. Presented at Google Summer of Code 2024 Conference (flash talk). Presenter: Sarrah Bastawala.
Saturday Morning NeuroSim, 9-14
Просмотров 6714 дней назад
Attendees: Bradly Alicea, Shubham Soni, Morgan Hough, Trevor Tanner, and Jesse Parent Update on lab activities and administration. The origins and intellectual connections of Active Inference. Active Inference and Reinforcement Learning. Computational Psychiatry discussion, current events and future trends. Discussion on research directions and open questions in NeuroAI: credit assignment, why ...
Cybernetics Reading Group, 9-12
Просмотров 4014 дней назад
Attendees: Bradly Alicea, Amanda Nelson, and Jesse Parent. Part II of the paper "Every Good Regulator Provides a World Model for Intelligent Systems", Section 2 (toy model examples) through Section 4 (discussion). Alicea et.al (2024). Every Good Regulator Provides a World Model for Intelligent Systems. www.researchgate.net/publication/381547888_Every_Good_Regulator_Provides_a_World_Model_for_In...
Cognition Futures Reading Group, 9-11
Просмотров 1314 дней назад
Attendees: Jesse Parent, Amanda Nelson, Morgan Hough, and Bradly Alicea. Admin work on projects and subprojects (via JoPro). Improving web presence for projects and integrating with Notion docs. Part I of the paper "Every Good Regulator Provides a World Model for Intelligent Systems", Abstract and Introduction. Alicea et.al (2024). Every Good Regulator Provides a World Model for Intelligent Sys...
Saturday Morning NeuroSim, 9-7
Просмотров 1221 день назад
Attendees: Trevor Tanner, Shubham Soni, Sarrah Bastawala, Jesse Parent, Morgan Hough, and Bradly Alicea Reviewing the current state of the lab's organoid initiatives. Overview of post-post-hexagonal CogSci event in Kendall Square (Cambridge) hosted by Aethos. Shubham Soni: Open-source sustainability as a Reinforcement Learning model using MESA. Reviewing the agent.py file. An overview of Jesse'...
Saturday Morning NeuroSim, 8-31
Просмотров 10Месяц назад
Attendees: Trevor Tanner, Shubham Soni, Jesse Parent, Morgan Hough, Lukas, and Bradly Alicea Google Summer of Code closing items on the lab Medium (medium.com/orel-group) and the Github repository (github.com/OREL-group/GSoC). Sarah’s work on LLMs for Open-source Sustainability using LLAMA and AutoCodeRover. Shubham’s work on Reinforcement Learning agents and the randomized condition for Open-s...
Saturday Morning NeuroSim, 8-24
Просмотров 4Месяц назад
Attendees: Jesse Parent, Bradly Alicea, Morgan Hough, Sarrah Bastawala, and Trevor Tanner Final presentation on LLMOSC (Large Language Models for Open-source Sustainability) project. discussion on upcoming Fall initiatives. BICA * AI 2024 overview. History and utility of Biologically-inspired Cognitive Architectures (BICAs). Discussion of grant proposals and recycling content. Open Dynamics of ...
GSoC/Open-source, 8-23
Просмотров 9Месяц назад
Attendees: Jesse Parent, Morgan Hough, Sarrah Bastawala, Himanshu Chougule, and Bradly Alicea. Open-source meeting in the Fall and Winter (plans to develop the time slot). Final presentation on LLMOSC (LLMs for Open-source Sustainability). Issue creation and a plausible stopping point. Discussion on Neurofedora and Linux fork development. A stack of contingencies, cutting-edge neuroscience, rel...
Saturday Morning NeuroSim, 8-10
Просмотров 13Месяц назад
Project updates, highlights of organoid initiatives. LLMOSC (open-source communities) and Automated issue creation. Cognitive Computational Neuroscience conference. Review of the paper "Space is a latent sequence". Theory of hippocampal-based spatial cognition. Development and reinforcement learning of spatial cognition and cognitive control. Model-based learning. References: Cognitive Computat...
Allostasis Machines @ BICA*AI 2024
Просмотров 1,4 тыс.Месяц назад
18-minute version, presented at BICA*AI 2024. Presenter: Bradly Alicea.
Saturday Morning Research Updates, 6-8
Просмотров 13 месяца назад
Saturday Morning Research Updates, 6-8
1:14:30 is when the Krakauer book - Complexity talk starts
RUclips just said to "hey nerrd, you wanna watch this??" And I'm like "yep" 😃
Very informative video. Thank you! One comment - LLMs do hallucinate in the sense of hallucination that you use (ruclips.net/video/hplkTym-vmI/видео.html). The completely arbitrary token production is rare with current SOTA models. Current models produce convincing text that is made up. This maps well to human hallucinations that can be thought of as predictions that never get grounded on reality so feel real.
Fascinating, didn't understand a word about it though
Interesting. Thanks. #4Percent
1:41:15 A Robot Walks Into a Bar discussion starts
Move over small world, welcome to the Universe ruclips.net/video/8XX0iRAN--8/видео.html
❤😊
as usual so very interesting, too packed!
Ah, thank you Bradly, I now see that you post your links and summary here on the youtub vid, Great thanks a lot, Had not realised this until today, so maybe my doc is redundant
a classic
Why did you get kicked out of the puck building?
Reminds me of anirban bandyopadhyay et. Al. And their nanobrain machine
Self assembling architectures are cool af
excited for paper!
Mirasol probinsyal bowl from dammam saudi arabia sex video sex video
This is aamazing! Thanks for sharing, OREL.
P r o m o S M 😘
woah!
was this supposed to be famous?
Amazing Explanation... From where can i join the community.
Join here: launchpass.com/orthogonal-research Our meetings are on Saturdays at 10am Eastern Time (North America). Join the Slack to get the link.
🎉
Looks cool! audio seems a bit low though :|
This was the outcome (in terms of code): github.com/OREL-group/DevelopmentalBVs/tree/main/MultiLearn. Also see the following paper: arxiv.org/abs/2003.07689
Hey thank you for your explication is very useful, but sorry that I ask this but, Could you share with me the link of the code or maybe the power point file about your explication? because i can't see well. Thank you again.
Good stuff!
45:00 Beginning of the "the learning curve for new fields of study" discussion arc
47:26 "grasping" comes up yet again, this time wrt grasping a sense of fullness of understanding a new field of study, or at least its general parameters.
37:30 Bradly begins his phase of onboarding / OREL tour and contribution philosophy
COOL
CogSci 2021 discussion becomes a little more in depth around 34:00
Great Presentation!
discussion of past instances of Princeton Envision starts ~ 45:00
Wasn't able to make it, interesting discussion!
7:38 - when I start introducing the conference and some of my personal background with it 11:37 - AI as a means to study ourselves - both intelligence and ethics 18:30 - The significance of familiarity with interdisciplinary teams and multiple perspectives on tech / society
:)
I never thought I’ll be able to recover my account, I have literally tried every video RUclips non worked out until I saw a comment here about hackersoft081 on IG and I hired his services in less than 30 minutes and I get my hack account back
Jowandahill
Thanks for this introduction. I'm doing my bachelor thesis on this soon
amazing stuff
Is there any further introduction of ptychography?
at the beginning when he says the lens he means the CCD.
is there supposed to be sound to this video?
No. Just an animation.