finding

active

finding:simulated-agent-achieves-perfect-performance-after-trial-12-following-bayesian-model-reduction-sleep-versus-14-trials-without-bmr

Simulated agent achieves perfect performance after trial 12 following Bayesian model reduction (sleep), versus ~14 trials without BMR.

Key result showing BMR dramatically accelerates rule learning in simulation

Source paper

extracted_from

Active Inference, Curiosity and Insight

(2017) · Karl Friston · Marco Lin · Chris Frith · Giovanni Pezzulo +2

Neighborhood — ranked by edge-count

Claims (1)

claim

Sleep implements Bayesian model reduction: non-REM sleep performs synaptic regression (complexity reduction) and REM sleep re-evaluates posteriors under the new reduced model.
associated_withsupports
Neurobiological interpretation linking sleep stages to distinct computational roles in model optimization

Related by similarity (8)

cosine ≥ 0.65 · no typed edge

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

In single-agent simulation of 32 trials, performance becomes perfect after trial 14 without Bayesian model reduction, with confidence increasing progressively.finding0.864
Baseline learning curve for pure epistemic learning without structure learning
Nearly all of 64 simulated agents attain 100% performance at around the 10th trial when allowed to perform abductive Bayesian model reduction after each trial.finding0.855
Group-level simulation result showing generalizability of BMR benefit across agents
Sleep (Bayesian model reduction) should improve performance on rule-learning tasks, with the prevalence of insight-dependent performance changes roughly doubling after nocturnal sleep.hypothesis0.829
Prediction consistent with Wagner et al. (2004) finding; extended to the active inference account of sleep
Bayesian model reduction after 12 trials correctly removes off-diagonal (redundant) parameters from the likelihood array, recovering the true contingency structure.finding0.777
Validation that BMR correctly identifies and prunes wrong connections in the likelihood mapping
A naive agent equipped with reduced priors from an experienced agent performs perfectly with maximum confidence from the first trial.finding0.765
Demonstration that model-level priors (not parameter-level knowledge) suffice for immediate transfer
In simulations, positive evidence threshold for Bayesian model reduction corresponds to ΔF ≤ −3, equivalent to odds ratio of exp(−3) ≈ 0.05 (reduced model ~20 times more likely than full model).finding0.759
Quantitative threshold used for accepting reduced models; linked to Bayes factor of ~20
3 of 64 simulated agents exhibited superstitious (incorrect) abduction, leading to persistently poor performance, demonstrating a trade-off between ampliative benefit and susceptibility to false insight.finding0.759
Demonstration of failure mode of abductive model reduction
Nocturnal sleep doubled the prevalence of insight-dependent performance improvements on a stimulus-response sequence task (Wagner et al., 2004).finding0.755
Empirical support for sleep as facilitator of structure learning; consistent with BMR account