Difference between revisions of "User:Louie Louiie"

From Sudo Room
Jump to navigation Jump to search
Line 1: Line 1:
<big>'''Artificial Super Organisms'''</big>
=Current Projects=
 
===Artificial Super Organisms (In Silico)===


'''Introduction'''
'''Introduction'''
Line 12: Line 14:


From the application perspective of this exercise, we can define each node as a micro-app written on the NodeJS platform, chosen for its decentralized concurrency structure and non-blocking I/O transaction capabilities<ref>[http://www.sosp.org/2001/papers/welsh.pdf Matt Welsh et al. – SEDA: An Architecture for Well-Conditioned, Scalable Internet Services (2001)]</ref>. Using a series of unique nodes as the initialization state, each node can be trained using neural networks/machine learning to do one of four tasks: worker, guard, librarians, oligarchy. A worker node accomplishes the learned tasks, unique to the purpose of application (eg: text analytics, image recognition, mathematical computation); the guard assigned to security enforcement functionality; librarians contain data; the oligarchy regulates policies to run the hive such as a queen in a colony. Each individual node can be replicated once the initial states are set. To accomplish a decentralized hive like structure for this exercise, individual node pathways may be configured in peer-to-peer (P2P) networks, forming a de facto blockchain as each node contains its own machine learned algorithm once a PGP layer is added<ref>[https://bitcoin.org/bitcoin.pdf Nakamoto, Satoshi (October 2008). "Bitcoin: A Peer-to-Peer Electronic Cash System"]</ref>.
From the application perspective of this exercise, we can define each node as a micro-app written on the NodeJS platform, chosen for its decentralized concurrency structure and non-blocking I/O transaction capabilities<ref>[http://www.sosp.org/2001/papers/welsh.pdf Matt Welsh et al. – SEDA: An Architecture for Well-Conditioned, Scalable Internet Services (2001)]</ref>. Using a series of unique nodes as the initialization state, each node can be trained using neural networks/machine learning to do one of four tasks: worker, guard, librarians, oligarchy. A worker node accomplishes the learned tasks, unique to the purpose of application (eg: text analytics, image recognition, mathematical computation); the guard assigned to security enforcement functionality; librarians contain data; the oligarchy regulates policies to run the hive such as a queen in a colony. Each individual node can be replicated once the initial states are set. To accomplish a decentralized hive like structure for this exercise, individual node pathways may be configured in peer-to-peer (P2P) networks, forming a de facto blockchain as each node contains its own machine learned algorithm once a PGP layer is added<ref>[https://bitcoin.org/bitcoin.pdf Nakamoto, Satoshi (October 2008). "Bitcoin: A Peer-to-Peer Electronic Cash System"]</ref>.
'''References'''
<references />


<br>
<br>
<hr>
<hr>


For more information, please contact me at: <br>
For more information, please contact me at: <br>

Revision as of 11:09, 26 September 2017

Current Projects

Artificial Super Organisms (In Silico)

Introduction

Super organisms such as bees, ants, and termites have superior efficiency in doing resilient, intensive, and complex tasks that have allowed them to adapt, evolve, and survive for a lengthy period of time[1]. This behavioral system can be artificially induced to create 'hive' like server architecture that can be programed to do agnostic functional computation independently with minimum human interference. Such that a human (or so we hope) positioned for the responsibility of owning the artificial hive would act as a caretaker moving the hive to its proper location, protecting it from black swans, etc.

Architecture

Based upon the laws of nature, the structure of such an artificial super organism would be replicable from a fractal based perspective. This architecture would look very much alike from quantized levels (network > server > application/db > code base). Each level having the same basic design, decentralized networking of nodes each with an algorithmic function of f(x) = y.

Technical

From the application perspective of this exercise, we can define each node as a micro-app written on the NodeJS platform, chosen for its decentralized concurrency structure and non-blocking I/O transaction capabilities[2]. Using a series of unique nodes as the initialization state, each node can be trained using neural networks/machine learning to do one of four tasks: worker, guard, librarians, oligarchy. A worker node accomplishes the learned tasks, unique to the purpose of application (eg: text analytics, image recognition, mathematical computation); the guard assigned to security enforcement functionality; librarians contain data; the oligarchy regulates policies to run the hive such as a queen in a colony. Each individual node can be replicated once the initial states are set. To accomplish a decentralized hive like structure for this exercise, individual node pathways may be configured in peer-to-peer (P2P) networks, forming a de facto blockchain as each node contains its own machine learned algorithm once a PGP layer is added[3].

References




For more information, please contact me at:
louis.huang@aegia.nu | +1.202.600.8398