Hi, I’m Ivan 👋

In my recent reading of Chapter 8 of “Artful Design”, I encountered a thought-provoking design principle: Design Principle 8.11: Design is the embodies conscience of technology, for technology has no conscience of its own. This principle resonates deeply with me as I stand at the crossroads of my academic journey, poised to step into the vast world beyond. As a final-year university student, my academic pursuits have oscillated between the rigorously technical - NLP, Signal Processing, and Deep Learning - and the creatively liberating courses at CCRMA, such as Fundamentals of Computer-Generated Sound and Music and AI....

In Chapter 7 of “Artful Design,” I recently came across a principle that deeply resonated with me: Principle 7.7: A little anonymity can go a long way. This principle took me down memory lane, particularly to my elementary school days when my Nintendo DSi was more than a gaming console; it was my portal to the internet. Before owning an iPod Touch or iPhone, I used the DSi’s web browser to explore sites like DSiCade and DSiPaint, designed specifically for DSi users....

Recently, I delved into Chapter 6 of Ge Wang’s “Artful Design,” and I couldn’t help but draw parallels with my own experiences in “CS 377G: Designing Serious Games,” a course I took last Spring under Christina Wodtke. This course remains one of my most memorable educational experiences, chiefly for its engaging content and the insightful principles it introduced. A particularly striking principle from Ge Wang’s book is: Principle 6.16: Games are perceived to be more accessible than instruments....

As I read this chapter, one of the principles that stuck out to me was Design Principle 5.5: Have Your Machine Learning – And the Human in the Loop! During my enrollment in Music and AI (CS470) last spring, I was profoundly impacted by this principle. The surge of generative AI tools, such as ChatGPT and Stable Diffusion, has stirred both excitement and trepidation in various sectors. Creatives, including artists and writers, express concerns about their place in an AI-dominated world....

This chapter prompted introspection about the projects I want to pursue in this course. I’m drawn to replicating the profound impact of teamLab exhibits I experienced in Japan. These exhibitions were not only visually striking but also highly interactive. One memory that stands out is an installation where rain-like graphics, emanating from the ceiling, would divert around a person standing underneath, emulating the deflection of raindrops. In another exhibit, projected flowers withered and died as visitors approached them....

In chapter three of “Artful Design,” a compelling discourse emerges on the intrinsic beauty of simplicity. It resonates with a truth many of us intuit but often overlook: that profound complexity can arise from the simplest of foundations. In particular, the principled was outlined as: Principle 3.5: Build complexity from simplicity. This tenet recalls notable emergent phenomena such as Conway’s Game of Life or Langton’s Ant. Both epitomize how simple rules can yield unpredictable and complex patterns....

In the annals of technological foresight, few have the prescience of Arthur C. Clarke. With an almost prophetic acuity, he envisioned the rise of the modern smartphone and its subsequent reverberation through the very fabric of human society. This notion brings to mind the insightful words of Peter Thiel, who once remarked on the power of inventing the future. When one juxtaposes these ideas, it becomes evident that Clarke’s prediction was not merely an observation but potentially a self-fulfilling prophecy....

Artful Design Chapter 1: Reading Response In this week’s reading of Artful Design, there was a particular principle that stood out to me: Principle 1.15: Design not only from needs – but from the values behind them. Most of the CS education that I have obtained at Stanford emphasizes the need to design computer systems and machine learning models that are meant to be efficient and reliable. Of course, it makes sense in context that we would want to design these things in such ways; who the hell would want to have a machine learning model that takes forever to classify a picture a plant into its correct species, only for that classification to wrong most of the time....

This is part three of my “Notes on ‘Non-linear PCA” blog article series, where I discuss various deep generative modeling techniques. In this section, I am going to discuss diffusion models. Introduction Diffusion models are different kinds of latent variable models compared to VAEs and GANs, and they take inspiration non-equilibrium statistical physics. The main idea behind them is that, given some dataset $\mathbf{X}$, we will repeatedly apply Gaussian noise to each data example $\mathbf{x}^{(i)}$ iteratively until the transformed data example is indistinguishable from noise sampled from the multivariate unit Gaussian....

This is part two of my “Notes on ‘Non-linear PCA” blog article series, where I discuss various deep generative modeling techniques. In this section, I am going to discuss GANs. Generative Adversarial Networks Generative Adversarial Networks (GANs) are a generative architecture based on a game-theoretical approach where two agents, a generator and a discriminator, compete to outperform the other in a zero-sum game. The discriminator’s main task is to differentiate between legitimate ground truth samples and fake samples produced by the generator network....