How my toddler is already a master electrical engineer

During the last few months of my master's degree, I would alternate my time looking up at the whiteboard to take notes and looking down at my belly to watch my baby move inside me. As he kicked, I sometimes wondered if he was paying attention to my professors. After all, babies can hear what's going on outside the womb.

Well, after several months of observation, I am pleased to report that Joseph must have been paying attention all along because he has already mastered several areas of electrical engineering. For example:

1. Basic Physics. In order to do any real electrical engineering, you've got to have your physics down. And Joseph does! Like a miniature Isaac Newton, he knows exactly how gravity works. It goes down. Everything, from balls to cheerios to cups of water, all follow the same basic rule. Along with this understanding of gravity, it appears he also has an understanding of the potential energy, U, of each of these objects. At least, that's my guess for why he announces, "Uh-oh" (ie U0) each time one of these experiments takes place.

2. Electricity and Magnetism. After you take your mechanics class, you've got to know your electricity and magnetism. And, judging by how much joy he gains from turning the lights on and off and pulling magnets on and off the fridge, I'd say he's pretty much there.

3. Stochastic Processes. Stochastic processes, or random processes, are at the heart of signal processing. Everything from compressed sensing to information theory depends on modeling the oh-so random world around us. And Joseph is certainly good at randomness, especially when it comes to food. Chicken might be okay one meal, but definitely not the next. The same goes for bread, strawberries, peas, milk, and basically every other food he eats. When you look at acceptable foods over time, the graph resembles white noise, a random signal. Naturally, the power spectral density of his food-acceptability curve is N0, or, in other words, "No!"

4. Coding Theory. As information theorist extraordinaire Claude Shannon proved, if you pick a probability of error and a rate of transmission less than capacity, there is an encoding and decoding scheme which will have error less than the chosen probability of error. In other words, if you're willing to work for it, you can get people to understand you. Right now, our communication takes place in a noisy channel because of a variety of factors, including a lack of fluency in each other's languages. However, Joseph is nothing if not persistent, and he overcomes the noisy channel by increasing the rate of transmission. That is, he knows how to follow me around saying, "Cracker? Cracker? Cracker?" in the hopes that I will someday decode his signals correctly.

5. Feature Detection. There are a variety of transforms out there used to identify features such as edges, corners, and even blobs. And somehow Joseph has managed to build those into his own brain. Take, for example, his use of the Radon transform. For weeks, every isolated line segment was a "Dih!" (stick), and upon this classification was usually picked up.

Nowadays, his transform of choice is the Hough transform, used to identify circles and spheres. Once recognized, it is classified as a "Baaaaahh!" (ball). From pumpkins to the moon in the sky to the letter "O," he has got the Hough transform down pat.

6. Detection and Estimation Theory. There is not an airplane nor helicopter nor bus that can go by without our notice. Joseph is a master at threshold detection. And then, once the object is located, he points to it the whole time it is in his sight. Must have an internal Kalman filter.

7. Machine Learning. Right now, Kevin's PhD uses a lot of machine learning, which is basically teaching a computer to learn from data. This is done by giving it lots and lots of data to help it recognize patterns. Since machine learning is about teaching a computer to think like a human, it's kind of silly to say that Joseph is good at it. I mean, of course he's good at learning to think like a human. He is one!

But, he is sure good at it. I mean, sometimes a computer needs tens of thousands of samples in several dimensions to figure things out. Then there are these huge matrices full of all this data that the computer has to somehow make sense of.  On the other hand, Joseph has probably only seen things a couple hundred times and has no matrix to place all the data (that we know of), and he's already got a lot down. Recognizing patterns? You bet! He knows that sunscreen + shoes = outside, for example. Following directions? He knows not only how to do what we ask, but also how to disregard us.

8. Electronics. Laptop, camera, cell phone, vacuum -- you name it, and he knows how it works. All the wiring and integrated circuit boards might still be a little too abstract for him, but he knows the most important part: If you turn it on, then it works.

So, as you can see, Joseph is well on his way to being a fantastic electrical engineer.

Or maybe he'll be a firefighter.

See also: How electrical engineering prepared me for motherhood