Yahoo (United States)

Keywords:Geoarchaeology, Palaeoenvironmental, Palaeo-Sediments, Landslide and Tectonic Movments, Roman Anchor, Greek harbour, Holocene Sea level change, Ras Hawala, Ras Alam El-Rum, Palaeo- tsunami, Graias Gony harbour, Egypt. Abstract:Graias Gony is submerged harbour in Ras Alam El-Rom area as a part of the NW coast of Egypt. Some geoarchaeological remains were discovered by Underwater Archaeology 20- 23rd October survey in the eastern side of Ras Alam El-Rum coastline on the NW coast of Egypt, East of Matruh city with extension about 16.9 km, SW Ras Alam El-Rom, in the same location of the above submerged harbours were described by another early writers.The geomorphological characteristics were studied for all locations but the coastline were changed from the Roman period up to recent time by coastal erosion and other geomorphic processes in addition to the Holocene sea level change.It was mentioned by Claude Ptolmee, who named it the Geographer Ptolemios, in his book in the year 150 AD in a list of ports in the Mediterranean Sea (Abd El Maguid. 2001). This port is one of the largest ports located on NW coast of Egypt and was recently named Rumaila harbour in reference to the type of coastal rock deposits spread in it along with The sandy sediments diminished greatly, which prompted the Bedouins to give it this name. The researcher found clear evidence showing sea level changes in this harbour, the most important of which are Notches and submerged platforms. I studied the geomorphological characteristics for their location, noting that coastal erosion, other geomorphic processes, and the Holocene Sea level change have altered the coastline from the Greek and Roman periods to the present. Alternating Quaternary limestone, part of an eroded carbonate coastal ridge, occupies the study area. We observed geomorphic coastal landforms along the shore, such as multi-level marine notches, platforms, and caves. These features were created at the Holocene relative sea level. We also observed solution microlandforms such as holes, channels, pits, and residual pinnacles. Sea water sculpted these features on carbonate rocks.

For a bucket test with a single criterion for success and a fixed number of samples or testing period, requiring a p-value less than a specified value of α for the success criterion produces statistical confidence at level 1 − α. For multiple criteria, a Bonferroni correction that partitions α among the criteria produces statistical confidence, at the cost of requiring lower p-values for each criterion. The same concept can be applied to decisions about early stopping, but that can lead to strict requirements for p-values. We show how to address that challenge by requiring criteria to be successful at multiple decision points.

Anaplastic thyroid cancer (ATC) is an aggressive, rare malignancy associated with rapid growth and metastasis, and a very poor prognosis. We investigated the clinical characteristics, survival outcomes and independent prognostic factors associated with anaplastic thyroid cancer. To assess to what extent the interaction between age and tumor stage affects mortality. A total of 622 patients diagnosed with anaplastic thyroid cancer, between 2010 and 2017 were enrolled in our study by retrieving data from the Surveillance, Epidemiology and End Results (SEER) database. We analyzed demographics, clinical characteristics, overall mortality (OM) and cancer specific mortality (CSM) of ATC. Variables with a P value < 0.1 were incorporated into the multivariate cox model to determine the independent prognostic factors. Furthermore, we analyzed the interaction between age and tumor stage on mortality. In the multivariate analyses, the divorced/separated population had a lower OM [hazard ratio (HR) = 0.63, 95%CI: 0.42-0.94, P < 0.05] and CSM (HR = 0.61, 95%CI: 0.40-0.92, P < 0.05). OM was higher in tumors with direct extension only (HR = 6.26, 95%CI: 1.29-30.42, P < 0.05) and tumors with distant spread (HR = 5.73, 95%CI: 1.34-24.51, P < 0.05). CSM was also higher in tumors with direct extension (HR = 5.05, 95%CI: 1.05-24.19, P < 0.05) and tumors with distant spread (HR = 4.57, 95%CI: 1.08-19.29, P < 0.05). Mortality was not adversely affected by lymph node involvement. OM was lower in patients who received radiation (HR = 0.66, 95%CI: 0.53-0.83, P < 0.01), chemotherapy (HR = 0.63, 95%CI: 0.50-0.79, P < 0.01) or surgery (HR = 0.53, 95%CI: 0.43-0.66, P < 0.01). CSM was also lower in patient who received radiation (HR = 0.64, 95%CI: 0.51-0.81, P < 0.01), chemotherapy (HR = 0.62, 95%CI: 0.50-0.78, P < 0.01) or surgery (HR = 0.51, 95%CI: 0.41-0.63, P < 0.01). There was no significant interaction between age and tumor stage that affected mortality. In this large US SEER database retrospective study, we found the mortality to be higher in advanced stage tumors with direct extension and distant metastasis. However, patients who received aggressive therapy showed a better overall survival. The aim of our study is to emphasize the importance of detecting ATC at an early stage and provide aggressive therapy to these patients. Since advanced stage ATC is associated with a dismal prognosis, we emphasize the need for randomized control trials and development of novel therapies that will be used to treat ATC.

We introduce a novel nonlinear Kalman filter that utilizes reparametrization gradients. The widely used parametric approximation is based on a jointly Gaussian assumption of the state-space model, which is in turn equivalent to minimizing an approximation to the Kullback-Leibler divergence. It is possible to obtain better approximations using the alpha divergence, but the resulting problem is substantially more complex. In this paper, we introduce an alternate formulation based on an energy function, which can be optimized instead of the alpha divergence. The optimization can be carried out using reparametrization gradients, a technique that has recently been utilized in a number of deep learning models.

Speech signals, typically sampled at rates in the tens of thousands per second, contain redundancies, evoking inefficiencies in sequence modeling. High-dimensional speech features such as spectrograms are often used as the input for the subsequent model. However, they can still be redundant. Recent investigations proposed the use of discrete speech units derived from self-supervised learning representations, which significantly compresses the size of speech data. Applying various methods, such as de-duplication and subword modeling, can further compress the speech sequence length. Hence, training time is significantly reduced while retaining notable performance. In this study, we undertake a comprehensive and systematic exploration into the application of discrete units within end-to-end speech processing models. Experiments on 12 automatic speech recognition, 3 speech translation, and 1 spoken language understanding corpora demonstrate that discrete units achieve reasonably good results in almost all the settings. We intend to release our configurations and trained models to foster future research efforts.

We design nearly-linear time numerical algorithms for the problem of multivariate multipoint evaluation over the fields of rational, real and complex numbers. We consider both exact and approximate versions of the algorithm. The input to the algorithms are (1) coefficients of an m-variate polynomial f with degree d in each variable, and (2) points a 1 , . . . , a N each of whose coordinate has value bounded by one and bit-complexity s. Approximate version: Given additionally an accuracy parameter t, the algorithm computes rational numbers β 1 , . . . , for all i, and has a running time of ((Nm + d m )(s + t)) 1+o( 1) for all m and all sufficiently large d. Exact version (when over rationals): Given additionally a bound c on the bit-complexity of all evaluations, the algorithm computes the rational numbers f (a 1 ), . . . , f (a N ), in time ((Nm + d m )(s + c)) 1+o(1) for all m and all sufficiently large d. .

Siamese-network-based self-supervised learning (SSL) suffers from slow convergence and instability in training. To alleviate this, we propose a framework to exploit intermediate self-supervisions in each stage of deep nets, called the Ladder Siamese Network. Our self-supervised losses encourage the intermediate layers to be consistent with different data augmentations to single samples, which facilitates training progress and enhances the discriminative ability of the intermediate layers themselves. While some existing work has already utilized multi-level self supervisions in SSL, ours is different in that 1) we reveal its usefulness with non-contrastive Siamese frameworks in both theoretical and empirical viewpoints, and 2) ours improves imagelevel classification, instance-level detection, and pixel-level segmentation simultaneously. Experiments show that the proposed framework can improve BYOL baselines by 1.0% points in ImageNet linear classification, 1.2% points in COCO detection, and 3.1% points in PASCAL VOC segmentation. In comparison with the state-of-the-art methods, our Ladder-based model achieves competitive and balanced performances in all tested benchmarks without causing large degradation in one.