From simple innate biases to complex visual concepts

78 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Toward an Integration of Deep Learning and Neuroscience (2016, methodology): “…delivered to an unsupervised learning network toward entities of interest to humans via an attentional process (Joscha Bach, personal communication), to bias hippocampal replay (Kumaran et al., 2016) or other aspects of memory access, or to directly train simple classiﬁers (Ullman et al., 2012).”
Unsupervised Learning of Sensorimotor Affordances by Stochastic Future Prediction (2018, background): “Unsupervised learning may play a critical role in how humans acquire these representations [46], but limited progress has been made in developing algorithms to learn sensorimotor relationships directly from sensory data.”
On the nature and origin of intuitive theories : learning, physics and psychology (2015, background): “I do not mean that some cues and heuristics are not real or useful(10), but it seems hopeless to think that a long list of cues and features can make something that has as output the reports of people upon seeing the Heider (10)As discussed in Chapters 2 and 3, these cues might act as lens that quic”