SOP for MS in Computer Science (Artificial Intelligence)

Hi all,

I'd appreciate your feedback on my statement of purpose for a computer science master's degree application. My declared area of concentration is artificial intelligence. The prompt is:


I squealed in excitement as I admired the jerky progress of my simulated creatures across the scrolling terrain on my laptop monitor. Their crudely-rendered bodies were composed of lines and squares, and as I watched, they took advantage of my "quirky" physics and learned to flip upward on the 2D hills. I had created the blocky creatures to demonstrate embodied cognition in automata, allowing them to interact with their environment and evolve an effective gait for navigating the dynamic terrain. Their lurching antics were easily more entertaining than Youtube videos of clumsy cats, and (under the guise of finding bugs in the physics I had written,) I proudly watched them shimmy, inch, lurch and wobble their way across the randomized terrain for the rest of the night.

The simulated creatures were part of a project for a particular professor whose open-ended assignments I loved to hijack to implement artificial intelligence techniques I had never worked with before. In another such project, I used his assignment of writing a Connect-4 AI player to familiarize myself with neural networks. Though it was an inappropriate task for a neural net, I was elated and surprised by the extent to which it could play with purpose.

Though it may seem naive to one more educated in the field than I, I am constantly surprised and excited by even the simplest new techniques that I learn.

A significant part of my motivation for studying artificial intelligence is simply that excitement. I could say that I'm interested in advancing the robotics industry, making better synthetic characters, or researching the nature of cognition-- and these things are true. But another part of my excitement about AI is entirely less lofty, exemplified by my glee in watching simulated creatures learn to walk: I am always exhilarated by writing a new type of agent and seeing it in action. I enjoy possessing the intellectual power to create things that are both dynamic and intelligent enough to keep me fascinated with not only the code and concepts behind them, but with their behavior itself.

Despite my child-like delight in the creation of intelligent agents, I am interested in more than fun and games with code. The sci-fi geek in me grew up with stories of body-swapping, clever androids, downloadable intelligence, and a myriad other technologies that seeded my curiosity about the nature of the mind. I am interested in studying the human mind as a biological machine, and the implications that may hold for artificial minds. I want to know how grey matter and our human selves intersect. I wonder about the nature of consciousness.

An academic mentor told me that the motivations of computer science students comprise a broad spectrum. There are those who love theoretical computer science, and those who romance the code. Then there are those who see the study of computer science as a way to learn how to formalize and implement solutions to a wide variety of problems outside of the field. I place myself firmly on this end of the spectrum. Formal theory may form a necessary foundation for my work and provide valuable insights and intuitions about how to go about solving a problem, but my passion lies in developing software to explore my curiosity about the nature of the mind and the limits of artificial intelligence. The question of whether or not we reach the point where I can sit down at a terminal and whip up a new soul may remain unanswered, but I continue to be fascinated by even the currently-possible in artificial intelligence.

I am most interested in the capabilities of AI to appear to or even be genuinely human-like, so I would be very excited to use my degree to start a career designing AI for video game characters or working on the development of interactive humanoid robots. However, since I would be happy working in many fields, I feel it is best to secure a more generalized education, explore varied topics, and keep my options open. Upon my acceptance to the computer science master's program, I would like to further my general education while pursuing my passion in the study and creation of intelligent agents. The Learning Agents research group is of particular interest, as I think that effective learning is a key part of human-like intelligence. I would also be excited to participate in the Neural Network research group, since bottom-up AI techniques like neural networks are so analogous to the human brain, and may someday aid in explaining how brain becomes mind.

My education so far has given me a good foundation in the field of artificial intelligence. However, I don't suspect such thoroughly researched and documented techniques as I've studied so far will continue to amaze me indefinitely. I wish to continue my studies in artificial intelligence at the University of Texas so that I can develop the new techniques that may lead to a better understanding of intelligence and the human mind.

==============

Grammar corrections are always welcome, but I'm more worried about the structure and if I have adequately answered the prompt. I did not talk about specific faculty members or research because my purpose is to take classes and I don't have a very strong desire to participate in a lot of research, or stay in academia (hence the master's.)

Thanks in advance for your time!
-Stacy

Hi Stacy,

I'd have to know about the output (what do most graduates do???) and curriculum of this program to appropriately advise you. Even if you don't plan to go into academia or research, you still may have to conduct ample research there--the prompt suggests that is the case. Moreover, some programs are more research-geared than others. Have you spoken to current students to investigate beyond the website? I usually find this is helpful. You can understand what the 'real expectations' of a program are on the inside.

My first instinct is that this SOP seems less grounded and strategic than most--you don't show program fit at all. If you intend to go into application development or the business realm, you also don't concretely paint that picture. And, you should at least do that. Overall, the language seems, as you pointed out, to romanticize CS too much--and I want to balance it with your concrete goals and interests.

Cheers,
Janson
Ivy Eyes Editing

Thanks, ivyeyes!

I'm pretty sure that this prompt is the same one that PhD applicants see, and for them it would be very important to explain how they fit in with the faculty and research. I am planning on doing a non-thesis Master's, and, as far as I understand it, should not be doing any research unless I happen to land a research assistantship, which I would do for the financial benefit. So I am a little at a loss as to how to show program fit.

I will definitely try and solidify in the essay what I plan to do after I get my degree. Honestly I would just like to take classes at a graduate level to explore this stuff! I don't have a very strong direction for my career because there are several areas where I can see myself being happy, and I tend to think that in this economy, chance-- whichever area I can actually find a job in-- is going to dictate where I go. In addition, I feel like without the actual education I'm trying to get, I can't make an informed decision about what I want to do. I'm not sure how to sugar-coat that for these programs that seem to assume you have your whole life planned out!

Thanks again.
-Stacy

Sounds good Stacy!

What do non-thesis Master's students do post-graduation? I think it would help you immensely to speak to some of these students (or maybe do some forum research).

My instinct is still to show what your long-term goals are, so that you really indicate how this program will help get you there. SOP's are very formulaic and strategic in this regard!

I can't tell you what the average student does, but I know I want to get SOME kind of a job. THAT is my goal. The problem is I really don't feel qualified to say that I want a career in some particular area without at least taking a class in it first. I don't think that that's ridiculous. The truth is: I am passionate about getting the opportunity to study several areas of AI, but lost if I have to declare right now what I want as a career. I mean, I've only had time to take a few relevant classes, none of which has really struck me with that AHA! moment. This uncertainty is the whole reason I want to go to graduate school. Maybe that makes me undesirable to schools, but I don't really know what else to do, because I really do want to continue my education!

Edit: Ivyeyes, I see now that you were probably trying to encourage me to look into this because it would be helpful to my essay. (It's late. Apologies for the rant.) But graduates of a program in computer science are useful for a huge variety of fields, and I'm not sure I'm going to gain any clarity from looking at where others have gone. I can see several possibilities, but like I said above, I feel like I'm not even qualified to say I want to do something when I have little to no experience in the field. I want to be educated!

I'm concerned about ivyeyes's comment that my language romanticizes CS too much-- believe it or not, this really is how I think and write, and I've heard from others that my writing style does come off aggrandizing and awkward at times. Are there any specific instances where I sound really ridiculous?

I have updated the essay, most notably with the second-to-last paragraph which attempts to add some concrete goals to the essay. I also mentioned a couple of research areas I would be interested in.

Thanks again for your time!
-Stacy

Okay, I stared at theintro for a long time, and I think my suggestion is this:

How about adding a very short sentence at the very beginning -- a sentence that starts with the words: Simulated creatures...

Or you could write:
I squealed in excitement while simulated creatures made j erky progress across the scrolling terrain of my laptop monitor. ----Ha ha, I think this is a cool first sentence...

I don't know, it's just my idea; not necessarily a great idea.

Thanks, Kevin. I had a friend read this last night and she also felt kind of lost in the first couple of sentences. I'll definitely try to make that first sentence more clear.

Thanks again!
-Stacy