In my first proto-board trial, I decided to recreate the ‘Color Mixing Lamp’ Arduino Beginner Project using a different RGB LED (along with some other modifications).
First I tested the circuit design on breadboard - showing the circuit, code and WS2812B RGB LED work together.
Using Fritzing, an electronics prototyping application, helped design the setup by quickly visualising component layouts.
After soldering each component and disconnecting rails (photo above).
Everything working including USB power from a Raspberry Pi!…
Week One: Online quiz on learn.gold Wekinator Assignment 1 on learn.gold Supervised Learning Ideas (forum link) Reflection: The aim of machine learning (ML) is to process an input in order to reach an output, without actually programming the decision-making stage (or the model in the pipeline) explicitly. My initial thoughts about this generalised ML schematic were that it seems extremely broad. Indeed, the input can be virtually any type of data; from single values to video streams.…
Documentation (photos and videos) of the next lot of Arduino projects completed for Physical Computing (IS53030B)…
07: Piezo keyboard w/ resistor ladder 08: Tilt switch timer 09: Motor control w/ battery and transistor 10: Dynamic motor control w/ integrated circuit 11: Alphanumeric LCD screen…
Stochastic diffusion search (SDS) is a swarm intelligence based algorithm that fundamentally differs to DFO (implemented last week) in the following ways:
Agents’ features define a hypothesis in the search space; a set of ‘microfeatures’ which together are a candidate for the solution In the ‘test phase’, random microfeatures in the hypotheses are used in partial evaluation instead of calculating the fitness of the whole Boolean (true/false) information is shared directly between random agents during the ‘diffusion phase’, unlike DFO agents’ results which is continuous When each negatively-evaluated agent communicates with a single random agent in the diffusion phase of SDS, it can either adopt the partner’s positively-evaluated hypothesis or (if the partner is also negative) a new random hypothesis is set.…
Documentation (photos, videos and GIFs) of the first half of the Arduino projects for Physical Computing (IS53030B)…
00: Setting up 01: Powered LED w/ switch and resistor 02: Digital out to multiple LEDs 03: Temperature sensor 04: Photoresistors (light sensors) 05: Servo control w/ potentiometer and capacitor 06: Piezo element (sound)…
The first swarm intelligence algorithm described was stochastic diffusion search (SDS). I heard from its author that current research on SDS may prove it to be Turing complete. If this is the case then any given algorithm could be executed within it - which is thought-provoking despite being presumably impractical for many computational tasks. SI may also be used to create new media tangential to academic research.
Algorithmic art has an obscure history but it seems to be an emerging movement.…
Swarm intelligence occurs in structures composed of simple ‘agents’ interacting at a local level, where intelligent behaviour emerges from the system as a whole. Foraging ants, flocking birds and swarming flies are examples of this behaviour that have each inspired different optimisation algorithms. Dispersive Flies Optimisation (DFO) is based on how flies hover over food sources in nature [1].
In each iteration of DFO the randomly generated agents can move based on the feature vector of two neighbors (left and right in the array), the best agent of the swarm and its own.…
The old saying “There ain’t no such thing as a free lunch” suggests that you can’t get something for nothing. It is also the namesake for the topic of this post: the “no free lunch” theorems (NFLT). NFLT revolves around the difficult problem of optimisation; finding the best result out of a set of alternatives.
As an example, I previously came across optimisation for computer vision in the task of separating an image’s background (also known as ‘background subtraction’).…