Generative Models
- PDF / 713,735 Bytes
- 19 Pages / 439.37 x 666.142 pts Page_size
- 57 Downloads / 221 Views
Generative Models Sim‑Hui Tee1 Received: 28 January 2019 / Accepted: 26 October 2020 © Springer Nature B.V. 2020
Abstract Generative models have been proposed as a new type of non-representational scientific models recently. A generative model is characterized with the capacity of producing new models on the basis of the existing one. The current accounts do not explain sufficiently the mechanism of the generative capacity of a generative model. I attempt to accomplish this task in this paper. I outline two antecedent accounts of generative models. I point out that both types of generative models function to generate new homogenous models in the sense that the latter is a straightforward derivative of the former, both of which share many similar features. Unfortunately, both accounts are implicit about the generative capacity of generative models. Using a case study, I articulate that a two-staged process of abstraction and idealization in modeling may contribute to the generative capacity of a scientific model. I also demonstrate that this two-staged process may go beyond the capacity of generating new homogenous models to generating new heterogeneous models.
1 Introduction Scientific models are widely regarded as representational tools in scientific research (Bolinska 2013; Frigg 2010; Giere 2010; Hughes 1997; Poznic 2017; Rohwer and Rice 2016; Woody 2004). Scientists construct models to represent the target of interest for the purpose of scientific explanation, understanding, and prediction. However, this representationalist view of scientific models has two limitations. First, not all models are representational (Grüne-Yanoff 2013; Luczak 2017; Suárez 1999). A model is non-representational when there is a lack of corresponding empirical content and of the applicability (Suárez 1999), or when the model is constructed to stand in for parallel worlds or possible processes (Grüne-Yanoff 2013; Sugden 2000). Second, the emphasis on the representational role of models imposes restrictive limitations on the epistemic value of models (Knuuttila 2005). Though we may * Sim‑Hui Tee [email protected]; [email protected] 1
Xiamen University Malaysia, Jalan Sunsuria, Bandar Sunsuria, 43900 Sepang, Selangor, Malaysia
13
Vol.:(0123456789)
S.-H. Tee
gain knowledge about the world through the representational relationship between our models and the target system, there are alternative ways through which we may also gain knowledge—such as through the interaction with the model by directly manipulating it (see Leonelli 2008; Toon 2011), or by resorting to the universality of holistically distorted models (Rice 2017a, b).1 In some circumstances, model manipulation may even enhance representation (Love and Travisano 2013). Turning from model representation to the activity of modeling may even broaden our view of the functions of models (Knuuttila 2005; Knuuttila and Boon 2011). Recently, generative models have been proposed as a new type of non-representational scientific models that are crucial in generating new mo
Data Loading...
