Local competition among neighboring neurons is common in biological neural networks (NNs). We apply the concept to gradient-based, backprop-trained artificial multilayer NNs. NNs with competing linear units tend to outperform those with non-competing nonlinear units, and avoid catastrophic forgetting when training sets change over time.

We investigate combinatorial lowness properties of sets of natural numbers (reals). The real A is super-low if A′ ≤tt ∅′, and A is jump-traceable if the values of {e}A(e) can be effectively approximated in a sense to be specified. We investigate those properties, in particular showing that super-lowness and jump-traceability coincide within the r.e. sets but none of the properties implies the o...

The following problem is addressed: given square matrices A and B, compute the smallest such that A + E and B + F have a common eigenvalue for some E, F with max(‖E‖2, ‖F‖2) ≤ . An algorithm to compute this quantity to any prescribed accuracy is presented, assuming that eigenvalues can be computed exactly.

‎in this paper we introduce mixed unitary cayley graph $m_{n}$‎ ‎$(n>1)$ and compute its eigenvalues‎. ‎we also compute the energy of‎ ‎$m_{n}$ for some $n$‎.

We use neuromorphic chips to perform arbitrary mathematical computations for the first time. Static and dynamic computations are realized with heterogeneous spiking silicon neurons by programming their weighted connections. Using 4K neurons with 16M feed-forward or recurrent synaptic connections, formed by 256K local arbors, we communicate a scalar stimulus, quadratically transform its value, a...

The roots of this problem go back to the beginnings of calculus and it is even sometimes called “Newton’s problem”. Historically, it has played a major role in the development of the theory of the integral. For example, it was Lebesgue’s primary motivation behind his theory of measure and integration. Indeed, the Lebesgue integral solves the primitive problem for the important special case when...

This paper focuses on complexity classes of partial functions that are computed in polynomial time with oracles in NPMV, the class of all multivalued partial functions that are computable nondeterministically in polynomial time. Concerning deterministic polynomial-time reducibilities, it is shown that 1. a multivalued partial function is polynomial-time computable with k adaptive queries to NPM...

A consistently specified halting function may be computed.

In prior work, we proposed the compute-andforward framework for sending linear combinations of messages to relays. In this note, we extend the notion of successive interference cancellation to the compute-and-forward setting. We find that once a relay has decoded a linear combination, it can mix it with its channel output to create a new effective channel output. The resulting effective channel...