PerceptionTime
My paper titled "Perception and Time in Artificial General Intelligence" (by rand3289)
distributAr
Framework for distributing user defined Spiking Neural Networks and other algorithms. (by rand3289)
PerceptionTime | distributAr | |
---|---|---|
58 | 13 | |
19 | 4 | |
- | - | |
4.7 | 4.1 | |
5 months ago | 30 days ago | |
C++ | ||
- | - |
The number of mentions indicates the total number of mentions that we've tracked plus the number of user suggested alternatives.
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
PerceptionTime
Posts with mentions or reviews of PerceptionTime.
We have used some of these posts to build our list of alternatives
and similar projects. The last one was on 2021-05-02.
-
Question for the experts
I would like to disagree. Spikes have everything to do with intelliegence. Perception to be exact. Here are my arguments: https://github.com/rand3289/PerceptionTime
-
[D] How do researchers generate idea of new machine learning architecture?
Over the next five years I kept thinking about a theory that would not use symbols during computation. Came up with some arguments but they weren't convincing enough. Now I live in my own world. All I got is this Idea that's not even an architecture but explanation why spiking NNs is the way to go: https://github.com/rand3289/PerceptionTime
-
Why Machine Learning might not be the way to AGI
Qualia can not be expressed symbolically. I have a theory that qualia can be expressed as a point in time though. A spike. This point in time is "formed" when an agent detected a change within itself. The change was caused by its environment. Here is more info: https://github.com/rand3289/PerceptionTime
-
The Chinese room argument, or Why Artificial Intelligence Doesn't Really Understand Anything
I would argue that spikes are points in time and are NOT symbolic in nature. They can be expressed as symbols (timestamps) however. Here is more info: https://github.com/rand3289/PerceptionTime
-
Does the eventual emergence of awareness in an AGI necessarily require a sensory apparatus?
Perception becomes subjective when there is a body. Here is more info: https://github.com/rand3289/PerceptionTime
-
Why Most Neuroscientific Theories of Consciousness Are Wrong
I completely agree! This is how I belive we can bring time into the model of consciousness: https://github.com/rand3289/PerceptionTime
-
Who are the big names working in the Neuroscience of Perception?
Here are some of my own ideas: https://github.com/rand3289/PerceptionTime
- Interesting article about inner workings of a neuron
- This turns everything I know about neurons upside down! Amazing new research.
- Why are AI models that require training data not seen as a viable route to AGI?
distributAr
Posts with mentions or reviews of distributAr.
We have used some of these posts to build our list of alternatives
and similar projects. The last one was on 2022-05-13.
-
Some speculation about the Winning Lottery Ticket Hypothesis
Here is the source: https://github.com/rand3289/distributAr
-
What is "missing" in robotics?
Software (AI) is the limiting factor. This is how I am working on that: https://github.com/rand3289/distributAr https://github.com/rand3289/PerceptionTime
-
Open-source AI?
And here is my framework for distributing spikes in a private IPv4 network: https://github.com/rand3289/distributAr
- The Hunt for the Ultimate AGI Algorithm
- Any way to connect multiple cheap GPUs to run some projects
-
Tiny framework for delivering spikes/events across the network.
#1: My tiny (2.5k lines of C++ code) framework for distributing SPIKING neural networks. | 0 comments #2: Introducing Pathways: A next-generation AI architecture | 0 comments #3: Neuron Bursts Can Mimic Famous AI Learning Strategy | 0 comments
I have a tiny (under 2000 lines of C++) opensource framework for distributing spiking ANNs. It could easily be adapted for other purposes. One distinguishing features is it has time synchronization among the nodes build in. Nodes can be connected manually or by drawing a graph and uploading it into the network. All you have to provide is a DLL that performs the actual computation and implements read() and write() methods. I hope this will be useful for someone. https://github.com/rand3289/distributAr
-
At an impasse with my research and need advice.
Do you have anything on paper? Any ideas? Surely you have something??? For example here is my framework for distributing spiking ANNs: https://github.com/rand3289/distributAr
-
I might have a safe AGI design, but I need help.
I noticed some similarity in what we are both working on. Although I do not have an algorithm, I have a "theory", a framework for distributing spikes (similar to your "architecture") and just like you, I am working on inexpensive sensor and actuator frameworks. I am not working on social or safety problems though... Here is the framework for distributing spikes: https://github.com/rand3289/distributAr The other things I've described are in the links in my earlier comments. I am a bit concerned about the fact you are not making implementation public. I hope you realize that Elon Musk created a company specifically for the purpose of sharing the technology so that no single entity posses it. My paper is really simple! I am stating that things that are alive process information by detecting changes in their internal state. Internal state can be meat or bits. It just happens that "detecting internal state change" is best described by a point in time (a timestamp or a spike). I call it a "perception mechanism" but it describes both sensory and non-sensory neurons.
- [R] Google's Launchpad Programming Framework Simplifies the Distributed Computation Learning Process
What are some alternatives?
When comparing PerceptionTime and distributAr you can also consider the following projects:
lim_lin