skip to main content
US FlagAn official website of the United States government
dot gov icon
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
https lock icon
Secure .gov websites use HTTPS
A lock ( lock ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

Explore Research Products in the PAR
It may take a few hours for recently added research products to appear in PAR search results.


Title: Universality and individuality in neural dynamics across large populations of recurrent networks
Many recent studies have employed task-based modeling with recurrent neural networks (RNNs) to infer the computational function of different brain regions. These models are often assessed by quantitatively comparing the low-dimensional neural dynamics of the model and the brain, for example using canonical correlation analysis (CCA). However, the nature of the detailed neurobiological inferences one can draw from such efforts remains elusive. For example, to what extent does training neural networks to solve simple tasks, prevalent in neuroscientific studies, uniquely determine the low-dimensional dynamics independent of neural architectures? Or alternatively, are the learned dynamics highly sensitive to different neural architectures? Knowing the answer to these questions has strong implications on whether and how to use task-based RNN modeling to understand brain dynamics. To address these foundational questions, we study populations of thousands of networks of commonly used RNN architectures trained to solve neuroscientifically motivated tasks and characterize their low-dimensional dynamics via CCA and nonlinear dynamical systems analysis. We find the geometry of the dynamics can be highly sensitive to different network architectures, and further find striking dissociations between geometric similarity as measured by CCA and network function, yielding a cautionary tale. Moreover, we find that while the geometry of neural dynamics can vary greatly across architectures, the underlying computational scaffold: the topological structure of fixed points, transitions between them, limit cycles, and linearized dynamics, often appears {\it universal} across all architectures. Overall, this analysis of universality and individuality across large populations of RNNs provides a much needed foundation for interpreting quantitative measures of dynamical similarity between RNN and brain dynamics.  more » « less
Award ID(s):
1845166
PAR ID:
10291268
Author(s) / Creator(s):
; ; ; ;
Date Published:
Journal Name:
Advances in neural information processing systems
Volume:
32
ISSN:
1049-5258
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ; ; ; ; (, Advances in neural information processing systems)
    null (Ed.)
    Recurrent neural networks (RNNs) are a widely used tool for modeling sequential data, yet they are often treated as inscrutable black boxes. Given a trained recurrent network, we would like to reverse engineer it--to obtain a quantitative, interpretable description of how it solves a particular task. Even for simple tasks, a detailed understanding of how recurrent networks work, or a prescription for how to develop such an understanding, remains elusive. In this work, we use tools from dynamical systems analysis to reverse engineer recurrent networks trained to perform sentiment classification, a foundational natural language processing task. Given a trained network, we find fixed points of the recurrent dynamics and linearize the nonlinear system around these fixed points. Despite their theoretical capacity to implement complex, high-dimensional computations, we find that trained networks converge to highly interpretable, low-dimensional representations. In particular, the topological structure of the fixed points and corresponding linearized dynamics reveal an approximate line attractor within the RNN, which we can use to quantitatively understand how the RNN solves the sentiment analysis task. Finally, we find this mechanism present across RNN architectures (including LSTMs, GRUs, and vanilla RNNs) trained on multiple datasets, suggesting that our findings are not unique to a particular architecture or dataset. Overall, these results demonstrate that surprisingly universal and human interpretable computations can arise across a range of recurrent networks. 
    more » « less
  2. ; ; ; (, Neural Computation)
    Abstract Recurrent neural networks (RNNs) are often used to model circuits in the brain and can solve a variety of difficult computational problems requiring memory, error correction, or selection (Hopfield, 1982; Maass et al., 2002; Maass, 2011). However, fully connected RNNs contrast structurally with their biological counterparts, which are extremely sparse (about 0.1%). Motivated by the neocortex, where neural connectivity is constrained by physical distance along cortical sheets and other synaptic wiring costs, we introduce locality masked RNNs (LM-RNNs) that use task-agnostic predetermined graphs with sparsity as low as 4%. We study LM-RNNs in a multitask learning setting relevant to cognitive systems neuroscience with a commonly used set of tasks, 20-Cog-tasks (Yang et al., 2019). We show through reductio ad absurdum that 20-Cog-tasks can be solved by a small pool of separated autapses that we can mechanistically analyze and understand. Thus, these tasks fall short of the goal of inducing complex recurrent dynamics and modular structure in RNNs. We next contribute a new cognitive multitask battery, Mod-Cog, consisting of up to 132 tasks that expands by about seven-fold the number of tasks and task complexity of 20-Cog-tasks. Importantly, while autapses can solve the simple 20-Cog-tasks, the expanded task set requires richer neural architectures and continuous attractor dynamics. On these tasks, we show that LM-RNNs with an optimal sparsity result in faster training and better data efficiency than fully connected networks. 
    more » « less
  3. ; ; ; (, Frontiers in Applied Mathematics and Statistics)
    Recurrent neural networks (RNNs) have been successfully applied to a variety of problems involving sequential data, but their optimization is sensitive to parameter initialization, architecture, and optimizer hyperparameters. Considering RNNs as dynamical systems, a natural way to capture stability, i.e., the growth and decay over long iterates, are the Lyapunov Exponents (LEs), which form the Lyapunov spectrum. The LEs have a bearing on stability of RNN training dynamics since forward propagation of information is related to the backward propagation of error gradients. LEs measure the asymptotic rates of expansion and contraction of non-linear system trajectories, and generalize stability analysis to the time-varying attractors structuring the non-autonomous dynamics of data-driven RNNs. As a tool to understand and exploit stability of training dynamics, the Lyapunov spectrum fills an existing gap between prescriptive mathematical approaches of limited scope and computationally-expensive empirical approaches. To leverage this tool, we implement an efficient way to compute LEs for RNNs during training, discuss the aspects specific to standard RNN architectures driven by typical sequential datasets, and show that the Lyapunov spectrum can serve as a robust readout of training stability across hyperparameters. With this exposition-oriented contribution, we hope to draw attention to this under-studied, but theoretically grounded tool for understanding training stability in RNNs. 
    more » « less
  4. ; ; ; (, NeurIPS 2023 Workshop NeurReps)
    Recurrent neural networks (RNNs) trained on a diverse ensemble of cognitive tasks, as described by Yang et al. (2019); Khona et al. (2023), have been shown to exhibit functional modularity, where neurons organize into discrete functional clusters, each specialized for specific shared computational subtasks. However, these RNNs do not demonstrate anatomical modularity, where these functionally specialized clusters also have a distinct spatial organization. This contrasts with the human brain which has both functional and anatomical modularity. Is there a way to train RNNs to make them more like brains in this regard? We apply a recent machine learning method, brain-inspired modular training (BIMT), to encourage neural connectivity to be local in space. Consequently, hidden neuron organization of the RNN forms spatial structures reminiscent of those of the brain: spatial clusters which correspond to functional clusters. Compared to standard L1 regularization and absence of regularization, BIMT exhibits superior performance by optimally balancing between task performance and sparsity. This balance is quantified both in terms of the number of active neurons and the cumulative wiring length. In addition to achieving brain-like organization in RNNs, our findings also suggest that BIMT holds promise for applications in neuromorphic computing and enhancing the interpretability of neural network architectures. 
    more » « less
  5. ; ; ; (, Annual International Conference of the IEEE Engineering in Medicine and Biology Society)
    Recent work characterized shifts in preparatory activity of the motor cortex during motor learning. The specific shift geometry during learning, washout, and relearning blocks was hypothesized to implement the acquisition, retention, and retrieval of motor memories. We sought to train recurrent neural network (RNN) models that could be used to study these motor learning phenomena. We built an environment for a curl field (CF) motor learning task and trained RNNs with reinforcement learning (RL) with novel regularization terms to perform behaviorally realistic reaching trajectories over the course of learning. Our choice of RL over supervised learning was motivated by the idea that motor adaptation, in the absence of demonstrations, is a process of reoptimization. We find these models, despite lack of supervision, reproduce many behavioral findings from monkey CF adaptation experiments. These models also captured key neurophysiological findings.We found that the model’s preparatory activity existed in a force-predictive subspace that remained stable across learning, washout, and relearning. Additionally, preparatory activity shifted uniformly, independently of the distance to the CF trained target. Finally, we found that the washout shift became more orthogonal to the learning shift, and hence more brain-like, when the RNNs were pretrained to have prior experience with CF dynamics. We argue the increased fit to neurophysiological recordings is driven by more generalizable and structured dynamical motifs in the model with more prior experience. This suggests that prior experience could organize preparatory neural activity underlying motor memory to have more orthogonal characteristics, by forming structured dynamical motifs in the motor cortex circuitry. 
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email