Modelling climate change in the context of the system that created it.
In the EU project WorldTrans we develop the System Dynamics model FRIDA to explore the possibilities for a transfer to a low-carbon future that “leaves no person behind” (the European Green Deal). Models that address such issues normally fall into two categories: Integrated Assessment models that study policies and behaviors that lead to greenhouse has emissions and Earth System climate models that study the change in climate that results from greenhouse gas emissions. With FRIDA we have coupled the two domains, and chosen to focus on the linkages and feedbacks of the system rather than the intricate details of the system (which normally makes it impossible to see the two domains in the same model).
Including the breadth of connections between the sectors in FRIDA has allowed us to capture unintuitive outcomes which arise from a more fully connected world-Earth system, for example that including climate damage functions reduces projections of warming and increases the projections of economic output.
System Dynamics is an approach to modelling that allows various voices (disciplines, methods, vocabularies) to be presented in the same quantitative framework. It therefore opens up for a wide range of research questions. In WorldTrans we have participants from very different fields, including psychology, economy, agriculture, meteorology and so forth.
For the curious CS student: SDGym is a potential for achieving responsible AI, using a System Dynamics model like FRIDA to understand the long-term impact of algorithmic interventions on society:
SDGym: Low-Code Reinforcement Learning Environments using System Dynamics Models
Understanding the long-term impact of algorithmic interventions on society is vital to achieving responsible AI. Traditional evaluation strategies often fall short due to the complex, adaptive and dynamic nature of society. While reinforcement learning (RL) can be a powerful approach for optimizing decisions in dynamic settings, the difficulty of realistic environment design remains a barrier to building robust agents that perform well in practical settings. To address this issue we tap into the field of system dynamics (SD) as a complementary method that incorporates collaborative simulation model specification practices. We introduce SDGym, a low-code library built on the OpenAI Gym framework which enables the generation of custom RL environments based on SD simulation models. Through a feasibility study we validate that well specified, rich RL environments can be generated from preexisting SD models and a few lines of configuration code. We demonstrate the capabilities of the SDGym environment using an SD model of the electric vehicle adoption problem. We compare two SD simulators, PySD and BPTK-Py for parity, and train a D4PG agent using the Acme framework to showcase learning and environment interaction. Our preliminary findings underscore the dual potential of SD to improve RL environment design and for RL to improve dynamic policy discovery within SD models. By open-sourcing SDGym, the intent is to galvanize further research and promote adoption across the SD and RL communities, thereby catalyzing collaboration in this emerging interdisciplinary space. (see:https://arxiv.org/abs/2310.12494)
As usual, pizza and refreshments will be served.