Ben Bogart


Ben Bogart explores the concept of “machine subjectivity” within the realm of artificial intelligence and machine learning. Bogart delves into the intricate relationship between human cognitive processes and autonomous machines. They challenge the conventional view of machine learning as merely objective statistical models, proposing instead that these systems can be seen as subjective entities capable of autonomous learning and decision-making.

Bogart discusses how machines, through unsupervised learning algorithms, categorize and interpret data, drawing imaginary boundaries that mimic human cognitive biases. This process highlights the subjective nature of machine perception, questioning the objectivity traditionally attributed to technological systems. Bogart uses visual examples from their artistic work, particularly their project that involves deconstructing and reconstructing cinematic frames, to illustrate how machines “perceive” and “imagine” based on the data they process.

Throughout the talk, Bogart emphasizes the importance of recognizing the subjective interpretations embedded within machine learning systems. They advocate for a deeper understanding of how these systems construct knowledge and the implications of their integration into societal frameworks, aiming to foster a dialogue that reassesses the interplay between human and machine cognition.

This presentation was part of the symposium ARTIFICIAL IMAGINATION which unites innovative artists engaged with emerging technologies. This focused on exploring and sharing their individual practices, experiences, and insights related to algorithms, artificial intelligence, and machine learning. It served as a platform for an enriching exchange of ideas between the artists and the audience, aiming to contribute a distinctive artistic viewpoint to the ongoing discussions about our evolving relationships with machine collaborators. Each session, including this one, highlighted how these technologies are being integrated and reflected in contemporary artistic processes, encouraging a broader understanding and appreciation of the creative potential of new digital tools.

Ben Bogart is a generative artist primarily working in installation and print whose practice is located at the intersection of art and science. His installations create content live in response to their sensed environment. Physical modelling, chaos, feedback systems, evolutionary algorithms, computer vision, and machine learning have been used to inform and engage in his creative process. Ben holds a Ph.D. in Interactive Arts and Technology from Simon Fraser University. In his Ph.D. research, he proposes an Integrative Theory of cognitive and neuro-biological mechanisms of perception, mental imagery, mind-wandering, and dreaming. This cognitive framework is manifest in a computational model and site-specific generative art installation: Dreaming Machine #3.

My hope really is that we can all consider the degree to which both our machines and our own mental processes are biased and not representations of reality, but really imaginary creative constructions in our minds. They just happen to be synchronized by the fact that we live in the same world.

if subjectivity is this interaction between sensation and imagination, what do I actually mean by imagination?

The Conceptual Framework of Subjective Machines

Ben Bogart

So a subjective machine, subjective the idea of subjective machines is a framework for thinking about objectivity and subjectivity, both in autonomous machines and also in brains that are in biological material. And this kind of follows from my PhD work, which was funded by a Social Science and Research Council of Canada on making a machine dream. And you can ask me more of that later. I’m not going to talk about it. So I would I’m going to get that definition out of the way right away and then talk more about details. But I define subjectivity as an interaction between sensation and imagination that forms a reinforcing pattern resulting in perception. Simple enough, there are two philosophical assumptions behind this definition of subjectivity. There’s probably more, but these these are two that I’m picking out and these ones are really important to me because they’re my own personal beliefs also. The first is that the world independent of cognition is the world that isn’t that we cannot conceive of. That is beyond senses and perception is actually there. But it is unevenly distributed and continuous. It’s not made up of discrete chunks of things, although some physicist would disagree. Second is following from Merleau-Ponty this idea that you can’t have a concept of what a subject is without already having a concept of what an object is that the idea of point of view and experience infers a relationship between two different sides. A thing to be perceived, but also a perceiver who is able to extract that thing from its world and think about it as something independent. So, if subjectivity is this interaction between sensation and imagination, what do I actually mean by imagination? I think imagination is the ability to construct things that aren’t a mirror of the outside world. So one way of thinking about that could be novelty generation, but I would define it as a cognitive process that facilitates that generation of internal subjective representations, in particular, including mental images. So what is sensation? Sensation allows information from the world as independent of cognition, that thing outside of us to implement those same subjective representations. So what’s the relationship between sensation, imagination and a machine? So those models sorry, those definitions of subjectivity, definitely apply, at least in my thinking, to biological agents. But maybe the idea of machine imagination is a little less clear.

Practical Implications and Artistic Integration

Ben Bogart

I would define machines sensation, for example, as the representational information patterns that are fed to a computer, and they’re often called the training data or a corpus of information, a body of information. And those are most often measurements of something could be atmospheric measurements or an image from a live camera or almost anything. The flipside of that is imagination. I think of the imagination of a machine as that as being implemented by UN supervised machine learning algorithms. So I’m not going to get into what that means very specifically beyond saying unsupervised machine learning algorithm is like a classifier. It decides that in this space of measurements, which things belong to cats, which things belong to dogs, which things are associated with maleness, which things are associated with femaleness? I think of that classification problem as the projection of boundaries into that underlying continuity that is independent of cognition. So here we have a whole bunch of points. This is about 10,000 individuals. The x axis, I believe, is height. The Y axis is weight. But I could have gotten those flipped. Doesn’t really matter for the purposes of what we’re going to talk about and this is classified or labeled data. So the white dots mean one thing and the black dots mean something else. So we might guess that since we’re talking about body mass, that there could be a gender thing going on here. We can see that the images at the very top here, there it is above a certain point. There are mostly white lines. Below a certain point, there’s mostly black points.

Machines and Us

Ben Bogart

By talking about machine subjectivity, my hope really is that we can all consider the degree to which both our machines and our own mental processes are biased and not representations of reality, but really imaginary creative constructions in our minds. They just happen to be synchronized by the fact that we live in the same world. We have that same underlying sensory information. And so maybe we should be considering the validity of all concepts in terms of this idea of a subjective point of view, a subjective point of view being a particular combination of similarity measures that is, deciding things are similar, are different abstract representations of things constructed from those similarities and differences. The contents of boundaries that are projected imaginary, and also that underlying information distribution, that thing that is independent of cognition, the actual numbers that informs that whole system. And I think by thinking about all of these things in the context of machines and ourselves, maybe we can get better at understanding diversity and the idea of where meaning fits in with validity of statistical models and also just computational ways of knowing the world.