Month: January 2022

Posted on

Controls

In a lot of driving games, unless you buy the special sticks, you only have two options: on and off. With the gas and brakes, you’re either on or off. You’re either hitting the steering wheel all the way to the right, the left, or off. There is no partial turn. Driving games made a huge leap forward when they allowed analog joysticks, variable turn wheels, pressure sensitive D-pads. When you could control the extent of a turn, feather the brake, control the gas.

Politics is regressing for the reasons driving games progressed. Digital controls are easier; analog is hard. ADCs must be built, tested, and employed.

I wonder if it has always been like this, and I’m coming not to a valley between political fortresses, but merely a valley between my perceptions of politics. Maybe my understanding of politics was always blurry, and I didn’t before see how high the fortresses were.

But that doesn’t matter. My objective is lower the fortresses and open the roads, clear the blockades and lower the drawbridges. I don’t see how to do it.

*An ADC is an Analog to Digital Converter. If you give it 4V out of 15, it translates that to 0100 and pulses on some sample frequency. Votes are digital: yes or no. Present/absent actually correspond to don’t know and undetermined voltage levels, which are a big thing in digital logic. I guess the closest we can come to ADCs in politics is voting sometimes one way, sometimes the other, and the frequency of those votes being significant.

Posted on

Features

I’m training a DL model on some data. Due to size constraints, I don’t feed the model the raw data. I preprocess and filter, reducing files down by a factor between a hundred and a thousand. So 10 million data points become 10 thousand or a 10×10,000 array, depending on specifics.

The upsides are:
1) Increased Accuracy
When I have a fair idea what I’m looking for.

2) Learning Something
A feature that works significantly better or worse than expected tells me something about the signal to noise ratio or inherent physics.

For example, I spend a lot of time in the time domain. I sample at varying speeds, often in the megahertz regime. Deep learning, and computers in general, only know what they are told, and that means if I’m using two samples as a features, V1 at time t1 and V2 at time t2, the model can’t explore relationships. It can weight them, but it cannot calculate V1/V2 and weight that, nor ln(V1) and ln(V2). It can if I create a layer that does that, but for exploratory purposes, it’s easier for me to calculate ln(V1) and use that as a feature than create a model that explores ln(V1).

What that means is as I feed features to this model, I’m exploring what works. When I find out that ln(V1) is a great feature but ln(V2) isn’t, that tells me something useful. That way I gain some insight into the functioning of the hidden physics, Plato’s puppet masters, and they’re as interesting as developing tools to expose them.

The downside is it’s hard to explore. It takes forever. This is why science is hard.

3) Reduced Computation Time
Depending on how I run the model, the cost of a calculation trends with the square of data size. Cost is a nebulous function of time and processing power. That factor of 1,000 from above, or even 100, often results in a 1/1,000,000 factor of computation time. That’s huge. That’s near the high-side, but reasonable time reduction factors of 1/100 total computation times are accessible.

I haven’t really started exploring this yet. I probably should.

Posted on

Attitude

The most useful thing you can do, of the things you can do, to improve your situation is typically improve your attitude. Redundancy for emphasis.

Posted on

Who do you think would win?

Me: Who do you think would win? Sauron, with the Ring of Power or the combined House of Amber? Oberon is still dead.

Answer: The Amberites are cosmic powers, so…

Me: Good call. To make it even, we’ll give the Amberites the nine rings.

Answer: …that really doesn’t make it better.

Me: =D