Catastrophic forgetting has been a significant problem hindering the deployment of deep learning algorithms in continual setting. Numerous methods have proposed to address catastrophic where an agent loses its generalization power old tasks while new tasks. We put forward alternative strategy handle with knowledge amalgamation (CFA), which learns student network from multiple heterogeneous teac...

We derive a recursion for the genus distributions of the graphs obtained by iteratively attaching a claw to the dipole D3. The minimum genus of the graphs in this sequence grows arbitrarily large. The families of graphs whose genus distributions have been calculated previously are either planar or almost planar, or they can be obtained by iterative single-vertex or single-edge amalgamation of s...

A graph product is the fundamental group of a graph of groups Amongst the simplest examples are HNN groups and free products with amalgamation. The conjugacy problem is solvable for recursively presented graph products with cyclic edge groups over finite graphs if the vertex groups have solvable conjugacy problem and the sets of cyclic generators in them are semicritical. For graph products ove...

We show that every free amalgamation class of finite structures with relations and (symmetric) partial functions is a Ramsey class when enriched by a free linear ordering of vertices. This is a common strengthening of the Nešetřil-Rödl Theorem and the second and third authors’ Ramsey theorem for finite models (that is, structures with both relations and functions). We also find subclasses with ...

We present a semantics for architectural specifications in Casl, including an extended static analysis compatible with modeltheoretic requirements. The main obstacle here is the lack of amalgamation for Casl models. To circumvent this problem, we extend the Casl logic by introducing enriched signatures, where subsort embeddings form a category rather than just a preorder. The extended model fun...

The amalgamation of leaf-labelled trees into a single supertree that displays each of the input trees is an important problem in classification. Clearly, there can be more than one (super) tree for a given set of input trees, in particular if a highly unresolved supertree exists. Here, we show (by explicit construction) that even if every supertree of a given collection of input trees is binary...