AI alignment as semi-permeable boundaries to contextualize optimization

Kurth-Nelson (2023) Kurth-Nelson and Sullivan et al. (2024) Gabriel et al. (2024) Heffner et al. (2025) Dohnány et al. (2025)

The brain spontaneously replays sets of memories in fast bursts to reason about their joint implications

Kurth-Nelson et al. (2016) Liu et al. (2019) Nour et al. (2021) Higgins et al. (2021) Liu et al. (2021) Kurth-Nelson et al. (2023) Schwartenbeck et al. (2023) Veselic et al. (2025)

Neural networks meta-learn how to learn faster, leading to brain-like properties

Wang et al. (2016) Wang and Kurth-Nelson et al. (2018) Ritter et al. (2018) Dasgupta et al. (2019)

In distributional reinforcement learning, dopamine circuits make many predictions simultaneously — ranging from pessimistic to optimistic

Dabney and Kurth-Nelson et al. (2020) Muller et al. (2024) Botvinick et al. (2025)