In this paper, we first show that any square-free monomial ideal in K[x_1, x_2, x_3, x_4, x_5] has the strong persistence property. Next, we provide a criterion for a minimal counterexample to the Conforti–Cornuéjols conjecture. Finally, we give a necessary and sufficient condition to determine the normally torsion-freeness of a linear combination of two normally torsion-free square-free monomial ideals.
Adversarial learning baselines for domain adaptation (DA) approaches in the context of semantic segmentation are under explored in semi-supervised framework. These baselines involve solely the available labeled target samples in the supervision loss. In this work, we propose to enhance their usefulness on both semantic segmentation and the single domain classifier neural networks. We design new training objective losses for cases when labeled target data behave as source samples or as real target samples. The underlying rationale is that considering the set of labeled target samples as part of source domain helps reducing the domain discrepancy and, hence, improves the contribution of the adversarial loss. To support our approach, we consider a complementary method that mixes source and labeled target data, then applies the same adaptation process. We further propose an unsupervised selection procedure using entropy to optimize the choice of labeled target samples for adaptation. We illustrate our findings through extensive experiments on the benchmarks GTA5, SYNTHIA, and Cityscapes. The empirical evaluation highlights competitive performance of our proposed approach.
The reliability of low-power wireless communications is compromised by the constant radio spectrum usage increase needed to support new applications. To allow reliable signal quality, more flexible countermeasures are needed to allow communication protocols despite crowded radio spectrum. This paper proposes to integrate Machine Learning (ML) in the internal processing of the Bluetooth Low Energy (BLE) radio. The objective is to predict the optimal gain of the Automatic Gain Control (AGC) index according to internal radio processing metrics produced during the previous received packets. The integration of a Bootstrap Aggregation Tree (BAgged Tree) in a simulated BLE radio shows reduction of 11% of the mean Packet Error Rate (PER) compared to the original performance.
