arXiv AI

Planning as Dynamics Relaxation: Hippocampal Recurrent Network Realizes Optimal Goal-Directed Navigation

arXiv Statistics ML
Oct 1

Learning to Plan from Random Exploration

arXiv:2609.38383v1 Announce Type: cross Abstract: Random exploration reveals how an environment can be traversed before a goal is specified. Can this experience support long-range planning without po...

By Deqian Kong, Guangyan Sun, Sheng Cheng, Sirui Xie, Bo Pang, Jianwen Xie, Tony Geng, Caiwen Ding, Ying Nian Wu
arXiv AI
1d ago

A Variational Latent Equilibrium for Learning in Neuronal Circuits

The paper proposes a biologically plausible framework to approximate backpropagation through time (BPTT) for learning complex temporal dependencies in neuronal circuits. It derives a time‑continuous equivalent of BPTT using a prospective energy function and introduces a learning rule for backward weights that ensures fully local space‑time dynamics, correcting gain errors inherent in previous approximations. This approach unifies and extends prior local, phase‑free credit assignment methods, offering a rigorous basis for spatiotemporal deep learning in the brain and guiding the design of physical circuits that implement these computations.

By Simon Brandt, Paul Haider, Walter Senn, Federico Benitez, Mihai A. Petrovici
arXiv AI
Sep 25

ELiSe: Efficient Learning of Sequences in Structured Recurrent Networks

The paper introduces ELiSe, a model that leverages cortical network scaffolds and dendritic compartments to learn complex non‑Markovian spatio‑temporal patterns using only local, always‑on, phase‑free synaptic plasticity. It demonstrates the model’s ability to acquire and replay intricate sequences, exemplified by a birdsong learning mock‑up, and shows robustness to external disturbances and flexibility in parameter settings.

By Laura Kriener, Kristin V\"olk, Ben von H\"unerbein, Federico Benitez, Walter Senn, Mihai A. Petrovici
arXiv AI
Sep 7

Compact Bellman-Grounded Cognitive Maps for Cost-Aware Navigation

Compact Bellman-Grounded Cognitive Maps (BCM) are introduced as a new method for cost-aware navigation that reuses a single learned map for different goals without per-goal retraining. BCM grounds the map in local edge costs using a self-supervised Bellman objective and a compact coordinate encoding, achieving near-optimal performance on weighted grids up to 1600 nodes with only a 5% mean gap to exact Dijkstra search. Its memory footprint grows sublinearly with graph size, outperforming connectivity-based spectral baselines and demonstrating scalability to complex environments.

By Yuzhe Han, Mingkun Xu, Yujie Wu
arXiv Machine Learning
Oct 9

Learning infinite context windows in recurrent architectures via spatial neural computing

The paper proposes a second‑order recurrent neural network that replaces traditional neuron‑to‑neuron communication with a spatially evolving field governed by discretized partial differential equations. This design creates a structured infinite‑order RNN where each state depends on its entire history, providing an unbounded receptive field while keeping the number of parameters fixed. The authors derive conditions for marginal stability that prevent vanishing or exploding gradients, and demonstrate that the architecture outperforms other recurrent models on long‑horizon benchmarks with fewer parameters.

By Aleix Salvador-Pomarol, Arthur N. Montanari, Earl K. Miller, Adilson E. Motter, Jorge Cort\'es
arXiv AI
Oct 1

Do Better Goal Representations Improve Goal-Conditioned Reinforcement Learning?

The paper investigates whether enhancing goal representations improves goal-conditioned reinforcement learning (GCRL) performance. By creating an exact temporal-distance goal representation in deterministic mazes and systematically degrading its geometric quality, the authors find that changes in goal representation have little effect on performance. In contrast, degrading the agent’s current state representation more than doubles failure rates, indicating that state representation is the critical bottleneck. The study further demonstrates that simple random Fourier positional encodings can significantly boost performance on challenging navigation tasks without additional map or objective modifications.

By Syed Nazmus Sakib, Abdul Monaf Chowdhury, Nafiul Haque, Shifat E Arman, Md Mehedi Hasan