Opinion TRENDS in Cognitive Sciences Vol.10 No.7 July 2006 Special Issue: Probabilistic models of cognition Theory-based Bayesian models of inductive learning and reasoning

Citations in the corpus

• Relational inductive biases, deep learning, and graph networks (2018, methodology): “When learning, we either fit new knowledge into our existing structured representations, or adjust the structure itself to better accommodate (and make use of) the new and the old (Tenenbaum et al., 2006; Griffiths et al., 2010; Ullman et al., 2017).”
• Deep Gaussian Processes (2012, background): “Most deep algorithms require a large amount of data to perform learning, however, we know that humans are able to perform inductive reasoning (equivalent to concept generalization) with only a few examples [Tenenbaum et al., 2006].”
• Taskonomy: Disentangling Task Transfer Learning (2018, background): “compositional modeling [33, 8, 11, 21, 53, 89, 87], homomorphic cryptography [40], lifelong learning [90, 13, 82, 81], functional maps [68], certain aspects of Bayesian inference and Dirichlet processes [52, 88, 87, 86, 35, 37], few-shot learning [78, 23, 22, 67, 83], transfer learning [72, 81, 27, ”
• Bayesian sampling in visual perception (2011, result): “The fact that low-level vision and perhaps low-level perception might involve sampling is particularly interesting in light of several other recent findings suggesting that higher-level cognitive tasks, like causal reasoning (41, 42) and decision-making (43), might also involve some form of sampling”
• Theory Acquisition as Stochastic Search (2010, unknown): “…recent work in computational cognitive modeling has attempted to formalize how intuitive theories are structured, used and acquired from experience (Tenenbaum, Griffiths, & Kemp, 2006), working broadly within a hierarchical Bayesian framework shown in Figure 1 (and explained in more detail below).”
• The Hamiltonian Brain: Efficient Probabilistic Inference with Excitatory-Inhibitory Neural Circuit Dynamics (2014, background): “Human behaviour is consistent with Bayesian inference in many sensory [1, 2, 3, 4], motor [5, 6] and cognitive [7, 8, 9] tasks.”
• Intuitive Theories (2020, unknown): “whereas only a small proportion of participants’ (24%) agreed that Sam caused the window to smash. While dependency theories have difficulty marking a difference between Frank and Sam (since the outcome counterfactually depended on both of their actions), process theories correctly predict that Fran”
• Synaptic sampling: A connection between PSP variability and uncertainty explains neurophysiological observations (2015, background): “In the laboratory, humans have been shown to use information about uncertainty when solving a variety of sensory [9, 10, 11, 12], motor [13, 14] and cognitive [15, 16, 17] tasks.”
• A glass-box interactive machine learning approach for solving NP-hard problems with the human-in-the-loop (2017, background): “solve computationally, but are able to exploit structural properties of the overall instance to improve parts of it (Knill and Pouget, 2004), (Tenenbaum et al., 2006).”
• On the nature and origin of intuitive theories : learning, physics and psychology (2015, background): “We build on much recent work addressing the ‘what’ question, which proposes to represent the content of children’s intuitive theories as probabilistic generative models defined over hierarchies of structured symbolic representations [177, 178, 90].”
• Probabilistic brains: knowns and unknowns (2013, background): “Recent studies 18 strongly suggest that humans use probabilistic inference to answer such questions.”
• Three Exemplars Allow at Least Some Linguistic Generalizations: Implications for Generalization Mechanisms and Constraints (2008, unknown): “This sketch can be put into Bayesian terms by saying that infants employ frequency-based priors to weight some generalizations more heavily in Bayesian hypothesis selection (Kemp, Perfors, & Tenenbaum, 2007; Tenenbaum, Griffiths, & Kemp, 2006 ).”
• A rational account of pedagogical reasoning: teaching by, and learning from, examples. (2014, unknown): “One strength of this approach is that it allows straightforward integration of different kinds of sampling assumptions through the use of generative models (Tenenbaum et al., 2006).”