concept
active
concept:kriegman-et-al-2019-automated-shapeshifting-for-function-recovery-in-damaged-robotsKriegman et al. 2019 Automated Shapeshifting for Function Recovery in Damaged Robots
Cited as evidence that hardware change can be faster than software change for injury recovery
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- Empirical inversion of the assumption that hardware changes are harder than software changes
- Evidence for blurring of embodied robot / non-embodied AI distinction through self-modeling
- Bongard robot self-modeling: robot discovers its shape, then reuses information after damage.finding0.732Key Artificial Life experiment illustrating remapping of self-model (Bongard et al. 2006).
- From Kriegman et al. (2021), a novel mode of reproduction never before seen.
- Claim formalizing the Anima Labs idea that transformers are effectively recurrent due to K/V stream.
- Related work studying capability of LLMs to subvert safety measures if severely misaligned
- Demonstrates bidirectional causal link: behavior manifold geometry can be recovered by optimizing in representation space.
- Key question raising doubt about whether incremental improvement can overcome the integrated anti-living system.