Interpretivism and AI

Engineering notes

Key topics: autonomous driving. See the paper for implementation details and experimental results.

Chinese explanation / 中文解读

中文解读待补充：本站会优先为端到端自动驾驶、BEV感知、3D目标检测、轨迹预测、路径规划、LiDAR感知等高价值论文补充中文说明。

Original abstract

The meaning of interpretivism is shifting. Its key characteristics — a focus on multiple meanings, the symbolic nature of interaction, the ongoing and performative making of social and material worlds, and the use qualitative methodologies — are being problematized by digitization generally and AI specifically. At the very least, digitization and, thus, quantification, underpins a significant portion of contemporary meaning-making processes and interactions. To understand social life, interpretivism must account for this quantification and, often in turn, needs to use at least some of these tools of quantification for the purpose of achieving the hermeneutic goal of understanding. In the process, the meaning of interpretivism is changing in ways that are often subliminal and thus not clearly accounted for. For example, computational grounded theory asserts that the goal of grounded theory has always been to “measure” meaning. Has it? Having been trained in grounded theory, this characterization does not resonate with me. While measurement may aid interpretation, the goals of these two activities do vary, and we should be clear about their differences. I will argue that in this era it is crucial that we be careful about how “interpretive” is used because it remains indispensable and is different from measurement. My goal is not to police boundaries, but rather to emphasize the specificity of what qualitative research means. This is necessary to understand how AI is changing research as well as how qualitative research can reconfigure the discussions regarding the meanings of AI.

Links and sources

• PDF from original source

Comments