Search

Cat qubits have emerged as a promising candidate for quantum computation due to their higher error-correction thresholds and low resource overheads. In existing literature, the detuning of the two-photon drive is assumed to be zero for implementing single and multi-qubit gates. We explore a modification of the Hamiltonian for a range of detuning and demonstrate that high fidelity single qubit gates can be performed even by proper parameter matching. We also analyze the CNOT gate in presence of an approximate detuning term and explain its fidelity improvements through Shortcut to Adiabaticity corrections., Comment: 11 pages, 6 figures

Superconducting metamaterial transmission lines implemented with lumped circuit elements can exhibit left-handed dispersion, where the group and phase velocity have opposite sign, in a frequency range relevant for superconducting artificial atoms. Forming such a metamaterial transmission line into a ring and coupling it to qubits at different points around the ring results in a multimode bus resonator with a compact footprint. Using flux-tunable qubits, we characterize and theoretically model the variation in the coupling strength between the two qubits and each of the ring resonator modes. Although the qubits have negligible direct coupling between them, their interactions with the multimode ring resonator result in both a transverse exchange coupling and a higher order $ZZ$ interaction between the qubits. As we vary the detuning between the qubits and their frequency relative to the ring resonator modes, we observe significant variations in both of these inter-qubit interactions, including zero crossings and changes of sign. The ability to modulate interaction terms such as the $ZZ$ scale between zero and large values for small changes in qubit frequency provides a promising pathway for implementing entangling gates in a system capable of hosting many qubits., Comment: 8 + 11 pages, 5 + 5 figures, 0 + 3 tables

In two-qubit gates activated by microwave pulses, by turning pulse on or off, the state of qubits are swapped between entangled or idle modes. In either mode, the presence of stray couplings makes qubits accumulate coherent phase error. However, the error rates in the two modes differ because qubits carry different stray coupling strengths in each mode; therefore, eliminating stray coupling from one mode does not remove it from the other. We propose to combine such a gate with a tunable coupler and show that both idle and entangled qubits can become free from stray couplings. This significantly increases the operational switch fidelity in quantum algorithms. We further propose a weakly-tunable qubit as an optimum coupler to bring the two modes parametrically near each other. This remarkably enhances the tuning process by reducing its leakage., Comment: 17 pages, 16 figures. arXiv admin note: substantial text overlap with arXiv:2202.05208

We propose a gate to switch superconducting qubit pairs in and out of a two-body interaction. This gate uses cross resonance driving on a tunable circuit with adjusted parameters and without accumulating residual ZZ interaction for idle and interacting qubits. It is imperative that this gate does not spread errors through the quantum register. Our detailed theoretical results show that these error-free modes do not necessarily require largely tunable circuits, such as magnetic modulation of qubits or couplers. We obtain the operational gate on weakly tuneable circuits as well and show that switching between them is remarkably fast., Comment: 15 pages, 15 figures

The conventional natural language processing approaches are not accustomed to the social media text due to colloquial discourse and non-homogeneous characteristics. Significantly, the language identification in a multilingual document is ascertained to be a preceding subtask in several information extraction applications such as information retrieval, named entity recognition, relation extraction, etc. The problem is often more challenging in code-mixed documents wherein foreign languages words are drawn into base language while framing the text. The word embeddings are powerful language modeling tools for representation of text documents useful in obtaining similarity between words or documents. We present a simple probabilistic approach for building efficient word embedding for code-mixed text and exemplifying it over language identification of Hindi-English short test messages scrapped from Twitter. We examine its efficacy for the classification task using bidirectional LSTMs and SVMs and observe its improved scores over various existing code-mixed embeddings