Sequential Learning and Catastrophic Forgetting in Differentiable Resistor Networks
For researchers in continual learning and physical computing, this provides a physically interpretable testbed to study forgetting, but the results are incremental, confirming known phenomena in a new domain.
The paper studies catastrophic forgetting in differentiable resistor networks, showing that forgetting is controlled by task conflict and adaptation degree, with a clear trade-off between forgetting and adaptation. Topology changes the balance, and forgetting is linked to localized conductance changes on high-current edges.
Differentiable physical networks provide a simple setting in which learning can be studied through the interaction between trainable parameters and physical equilibrium constraints. We investigate sequential learning in differentiable resistor networks governed by Kirchhoff's laws. Although individual input--output mappings can be learned by gradient-based adjustment of edge conductances, sequential training on conflicting tasks produces catastrophic forgetting. We show that forgetting is controlled by task conflict and by the degree of adaptation to the new task. Uniform anchoring and normalised gradient-weighted anchoring reduce forgetting only by increasing the final loss on the new task, giving a clear forgetting--adaptation trade-off. We also show that forgetting is associated with localised conductance changes on high-current edges, giving a physical interpretation as reconfiguration of dominant transport pathways. Broader random-task ensembles show that the strongest forgetting occurs when the second task reverses the output ordering imposed by the first task. Finally, comparisons across Erdős--Rényi, small-world, scale-free, and random-geometric graph ensembles show that topology changes the forgetting--adaptation balance. These results position differentiable resistor networks as compact, physically interpretable testbeds for studying continual learning in tunable matter.