arXiv subjects

Randall C. O'Reilly

Publications and source records attributed to Randall C. O'Reilly.

15 recordsLinked to original sources

arXiv2026

This is how the Neocortex Learns

A sufficient account of how the neocortex learns must meet three criteria: 1. Computationally, it must approximate a powerful, general-purpose learning algorithm known to scale to human-level intelligence; 2. Algorithmically, it must be implementable using known, well-established neural circuits within the neocortex and associated brain structures; 3. Implementationally, there must be a detailed account for how all of the algorithmic mechanisms actually function at a neurochemical level. At present, there is only one framework that meets all of these criteria: error-driven predictive learning via temporal derivatives, driven by corticothalamic circuits, based on competitive kinase synaptic plasticity induction mechanisms. This has been implemented in the Axon neural simulation framework using spiking neurons, and demonstrated to learn across a wide range of challenging cognitively motivated tasks.

q-bio.NC↗

arXiv2025

Flexible Prefrontal Control over Hippocampal Episodic Memory for Goal-Directed Generalization

Many tasks require flexibly modifying perception and behavior based on current goals. Humans can retrieve episodic memories from days to years ago, using them to contextualize and generalize behaviors across novel but structurally related situations. The brain's ability to control episodic memories based on task demands is often attributed to interactions between the prefrontal cortex (PFC) and hippocampus (HPC). We propose a reinforcement learning model that incorporates a PFC-HPC interaction mechanism for goal-directed generalization. In our model, the PFC learns to generate query-key representations to encode and retrieve goal-relevant episodic memories, modulating HPC memories top-down based on current task demands. Moreover, the PFC adapts its encoding and retrieval strategies dynamically when faced with multiple goals presented in a blocked, rather than interleaved, manner. Our results show that: (1) combining working memory with selectively retrieved episodic memory allows transfer of decisions among similar environments or situations, (2) top-down control from PFC over HPC improves learning of arbitrary structural associations between events for generalization to novel environments compared to a bottom-up sensory-driven approach, and (3) the PFC encodes generalizable representations during both encoding and retrieval of goal-relevant memories, whereas the HPC exhibits event-specific representations. Together, these findings highlight the importance of goal-directed prefrontal control over hippocampal episodic memory for decision-making in novel situations and suggest a computational mechanism by which PFC-HPC interactions enable flexible behavior.

cs.NE↗

arXiv2023

Learning Representations that Support Extrapolation

Extrapolation -- the ability to make inferences that go beyond the scope of one's experiences -- is a hallmark of human intelligence. By contrast, the generalization exhibited by contemporary neural network algorithms is largely limited to interpolation between data points in their training corpora. In this paper, we consider the challenge of learning representations that support extrapolation. We introduce a novel visual analogy benchmark that allows the graded evaluation of extrapolation as a function of distance from the convex domain defined by the training data. We also introduce a simple technique, temporal context normalization, that encourages representations that emphasize the relations between objects. We find that this technique enables a significant improvement in the ability to extrapolate, considerably outperforming a number of competitive techniques.

cs.CV↗

arXiv2022

A Neural Network Model of Continual Learning with Cognitive Control

Neural networks struggle in continual learning settings from catastrophic forgetting: when trials are blocked, new learning can overwrite the learning from previous blocks. Humans learn effectively in these settings, in some cases even showing an advantage of blocking, suggesting the brain contains mechanisms to overcome this problem. Here, we build on previous work and show that neural networks equipped with a mechanism for cognitive control do not exhibit catastrophic forgetting when trials are blocked. We further show an advantage of blocking over interleaving when there is a bias for active maintenance in the control signal, implying a tradeoff between maintenance and the strength of control. Analyses of map-like representations learned by the networks provided additional insights into these mechanisms. Our work highlights the potential of cognitive control to aid continual learning in neural networks, and offers an explanation for the advantage of blocking that has been observed in humans.

q-bio.NC↗

arXiv2021

Locally Learned Synaptic Dropout for Complete Bayesian Inference

The Bayesian brain hypothesis postulates that the brain accurately operates on statistical distributions according to Bayes' theorem. The random failure of presynaptic vesicles to release neurotransmitters may allow the brain to sample from posterior distributions of network parameters, interpreted as epistemic uncertainty. It has not been shown previously how random failures might allow networks to sample from observed distributions, also known as aleatoric or residual uncertainty. Sampling from both distributions enables probabilistic inference, efficient search, and creative or generative problem solving. We demonstrate that under a population-code based interpretation of neural activity, both types of distribution can be represented and sampled with synaptic failure alone. We first define a biologically constrained neural network and sampling scheme based on synaptic failure and lateral inhibition. Within this framework, we derive drop-out based epistemic uncertainty, then prove an analytic mapping from synaptic efficacy to release probability that allows networks to sample from arbitrary, learned distributions represented by a receiving layer. Second, our result leads to a local learning rule by which synapses adapt their release probabilities. Our result demonstrates complete Bayesian inference, related to the variational learning method of dropout, in a biologically constrained network using only locally-learned synaptic failure rates.

q-bio.NC↗

arXiv2021

Statistical Learning in Speech: A Biologically Based Predictive Learning Model

Infants, adults, non-human primates and non-primates all learn patterns implicitly, and they do so across modalities. The biological evidence supports the hypothesis that the mechanism for this learning is general but computationally local. We hypothesize that the mechanism itself is predictive error-driven learning. We build on recent work that advanced a biologically plausible model of error backpropagation learning which proposes that higher order thalamic nuclei provide a locale for a temporal difference between top-down predictions and an actual event outcome. Our neural network based on that work also models the auditory cortex hierarchy of core, belt and parabelt and the caudal-rostral axis within regions. We simulated two studies showing statistical learning in infants, a seminal study using synthesized speech and a more recent study using human speech. Before simulating these studies the network was trained on spoken sentences from the TIMIT corpus to emulate infant's experience listening to random speech. The implemented neural network, learning only by predicting the next brief speech segment, learned in both simulations to predict in-word syllables better than next-word syllables showing that prediction could be the basis for word segmentation and thus statistical learning.

q-bio.NC↗

arXiv2021

The Structure of Systematicity in the Brain

A hallmark of human intelligence is the ability to adapt to new situations, by applying learned rules to new content (systematicity) and thereby enabling an open-ended number of inferences and actions (generativity). Here, we propose that the human brain accomplishes these feats through pathways in the parietal cortex that encode the abstract structure of space, events, and tasks, and pathways in the temporal cortex that encode information about specific people, places, and things (content). Recent neural network models show how the separation of structure and content might emerge through a combination of architectural biases and learning, and these networks show dramatic improvements in the ability to capture systematic, generative behavior. We close by considering how the hippocampal formation may form integrative memories that enable rapid learning of new structure and content representations.

q-bio.NC↗

arXiv2021

Complementary Structure-Learning Neural Networks for Relational Reasoning

The neural mechanisms supporting flexible relational inferences, especially in novel situations, are a major focus of current research. In the complementary learning systems framework, pattern separation in the hippocampus allows rapid learning in novel environments, while slower learning in neocortex accumulates small weight changes to extract systematic structure from well-learned environments. In this work, we adapt this framework to a task from a recent fMRI experiment where novel transitive inferences must be made according to implicit relational structure. We show that computational models capturing the basic cognitive properties of these two systems can explain relational transitive inferences in both familiar and novel environments, and reproduce key phenomena observed in the fMRI experiment.

q-bio.NC↗

arXiv2021

Deep Predictive Learning in Neocortex and Pulvinar

How do humans learn from raw sensory experience? Throughout life, but most obviously in infancy, we learn without explicit instruction. We propose a detailed biological mechanism for the widely-embraced idea that learning is based on the differences between predictions and actual outcomes (i.e., predictive error-driven learning). Specifically, numerous weak projections into the pulvinar nucleus of the thalamus generate top-down predictions, and sparse, focal driver inputs from lower areas supply the actual outcome, originating in layer 5 intrinsic bursting (5IB) neurons. Thus, the outcome is only briefly activated, roughly every 100 msec (i.e., 10 Hz, alpha), resulting in a temporal difference error signal, which drives local synaptic changes throughout the neocortex, resulting in a biologically-plausible form of error backpropagation learning. We implemented these mechanisms in a large-scale model of the visual system, and found that the simulated inferotemporal (IT) pathway learns to systematically categorize 3D objects according to invariant shape properties, based solely on predictive learning from raw visual inputs. These categories match human judgments on the same stimuli, and are consistent with neural representations in IT cortex in primates.

q-bio.NC↗

arXiv2019

Compositional generalization in a deep seq2seq model by separating syntax and semantics

Standard methods in deep learning for natural language processing fail to capture the compositional structure of human language that allows for systematic generalization outside of the training distribution. However, human learners readily generalize in this way, e.g. by applying known grammatical rules to novel words. Inspired by work in neuroscience suggesting separate brain systems for syntactic and semantic processing, we implement a modification to standard approaches in neural machine translation, imposing an analogous separation. The novel model, which we call Syntactic Attention, substantially outperforms standard methods in deep learning on the SCAN dataset, a compositional generalization task, without any hand-engineered features or additional supervision. Our work suggests that separating syntactic from semantic learning may be a useful heuristic for capturing compositional structure.

cs.LG↗

arXiv2017

Deep Predictive Learning: A Comprehensive Model of Three Visual Streams

How does the neocortex learn and develop the foundations of all our high-level cognitive abilities? We present a comprehensive framework spanning biological, computational, and cognitive levels, with a clear theoretical continuity between levels, providing a coherent answer directly supported by extensive data at each level. Learning is based on making predictions about what the senses will report at 100 msec (alpha frequency) intervals, and adapting synaptic weights to improve prediction accuracy. The pulvinar nucleus of the thalamus serves as a projection screen upon which predictions are generated, through deep-layer 6 corticothalamic inputs from multiple brain areas and levels of abstraction. The sparse driving inputs from layer 5 intrinsic bursting neurons provide the target signal, and the temporal difference between it and the prediction reverberates throughout the cortex, driving synaptic changes that approximate error backpropagation, using only local activation signals in equations derived directly from a detailed biophysical model. In vision, predictive learning requires a carefully-organized developmental progression and anatomical organization of three pathways (What, Where, and What * Where), according to two central principles: top-down input from compact, high-level, abstract representations is essential for accurate prediction of low-level sensory inputs; and the collective, low-level prediction error must be progressively and opportunistically partitioned to enable extraction of separable factors that drive the learning of further high-level abstractions. Our model self-organized systematic invariant object representations of 100 different objects from simple movies, accounts for a wide range of data, and makes many testable predictions.

q-bio.NC↗

arXiv2015

The cerebellum could solve the motor error problem through error increase prediction

We present a cerebellar architecture with two main characteristics. The first one is that complex spikes respond to increases in sensory errors. The second one is that cerebellar modules associate particular contexts where errors have increased in the past with corrective commands that stop the increase in error. We analyze our architecture formally and computationally for the case of reaching in a 3D environment. In the case of motor control, we show that there are synergies of this architecture with the Equilibrium-Point hypothesis, leading to novel ways to solve the motor error problem. In particular, the presence of desired equilibrium lengths for muscles provides a way to know when the error is increasing, and which corrections to apply. In the context of Threshold Control Theory and Perceptual Control Theory we show how to extend our model so it implements anticipative corrections in cascade control systems that span from muscle contractions to cognitive operations.

q-bio.NC↗

arXiv2014

Learning Through Time in the Thalamocortical Loops

We present a comprehensive, novel framework for understanding how the neocortex, including the thalamocortical loops through the deep layers, can support a temporal context representation in the service of predictive learning. Many have argued that predictive learning provides a compelling, powerful source of learning signals to drive the development of human intelligence: if we constantly predict what will happen next, and learn based on the discrepancies from our predictions (error-driven learning), then we can learn to improve our predictions by developing internal representations that capture the regularities of the environment (e.g., physical laws governing the time-evolution of object motions). Our version of this idea builds upon existing work with simple recurrent networks (SRN's), which have a discretely-updated temporal context representations that are a direct copy of the prior internal state representation. We argue that this discretization of temporal context updating has a number of important computational and functional advantages, and further show how the strong alpha-frequency (10hz, 100ms cycle time) oscillations in the posterior neocortex could reflect this temporal context updating. We examine a wide range of data from biology to behavior through the lens of this LeabraTI model, and find that it provides a unified account of a number of otherwise disconnected findings, all of which converge to support this new model of neocortical learning and processing. We describe an implemented model showing how predictive learning of tumbling object trajectories can facilitate object recognition with cluttered backgrounds.

q-bio.NC↗

arXiv2014

Goal-Driven Cognition in the Brain: A Computational Framework

Current theoretical and computational models of dopamine-based reinforcement learning are largely rooted in the classical behaviorist tradition, and envision the organism as a purely reactive recipient of rewards and punishments, with resulting behavior that essentially reflects the sum of this reinforcement history. This framework is missing some fundamental features of the affective nervous system, most importantly, the central role of goals in driving and organizing behavior in a teleological manner. Even when goal-directed behaviors are considered in current frameworks, they are typically conceived of as arising in reaction to the environment, rather than being in place from the start. We hypothesize that goal-driven cognition is primary, and organized into two discrete phases: goal selection and goal engaged, which each have a substantially different effective value function. This dichotomy can potentially explain a wide range of phenomena, playing a central role in many clinical disorders, such as depression, OCD, ADHD, and PTSD, and providing a sensible account of the detailed biology and function of the dopamine system and larger limbic system, including critical ventral and medial prefrontal cortex. Computationally, reasoning backward from active goals to action selection is more tractable than projecting alternative action choices forward to compute possible outcomes. An explicit computational model of these brain areas and their function in this goal-driven framework is described, as are numerous testable predictions from this framework.

q-bio.NC↗

arXiv2011

Surely You Must All be Joking: An Outsider's Critique of Quantum Physics

A critique of the state of current quantum theory in physics is presented, based on a perspective outside the normal physics training. From this perspective, the acceptance of quantum nonlocality seems unwarranted, and the fundamental assumptions that give rise to it in the first place seem questionable, based on the current status of the quantum theory of light. The relevant data can instead be accounted for using physically motivated local models, based on detailed properties of the experimental setups. The semiclassical approach, particularly in the form of the fully coupled Maxwell-Dirac equations with a pure wave ontology, seems to provide a satisfying, local, paradox-free physical model of the quantum world, that appears consistent with known phenomena. It is unclear why this approach is not pursued more vigorously in the field, given its clear potential to resolve all the conundrums that have perplexed generations of physicists.

physics.gen-ph↗