Multivariate time series forecasting (MTSF) often faces challenges from missing variables, which hinder conventional spatial-temporal graph neural networks in modeling inter-variable correlations. While GinAR addresses variable missing using attention-based imputation and adaptive graph learning for the first time, it lacks interpretability and fails to capture more latent temporal patterns due to its simple recursive units (RUs). To overcome these limitations, we propose the Interpretable Bidirectional-modeling Network (IBN), integrating Uncertainty-Aware Interpolation (UAI) and Gaussian kernel-based Graph Convolution (GGCN). IBN estimates the uncertainty of reconstructed values using MC Dropout and applies an uncertainty-weighted strategy to mitigate high-risk reconstructions. GGCN explicitly models spatial correlations among variables, while a bidirectional RU enhances temporal dependency modeling. Extensive experiments show that IBN achieves state-of-the-art forecasting performance under various missing-rate scenarios, providing a more reliable and interpretable framework for MTSF with missing variables. Code is available at: https://github.com/zhangth1211/NICLab-IBN.

While humanoid robots that can do box jumps and back flips get more attention than enterprise networking tech--after all, if you're not in the IT business, you have no clue what SD-WAN is--2020 is poised to be the year when trends become reality. One of those much-hyped trends is 5G. But it's more likely to be used by consumers than enterprises in the coming year. So what does that mean for your enterprise network? "In 2020, Wi-Fi 6 will enter the enterprise, through the employee door and through enterprise access point refreshes. Indeed, Cisco projects more people will be using Wi-Fi 6 in 2020 than 5G. In part, that's because 5G service still won't provide "the ultra-high speed connectivity that we have been promised or that we will see in future years," Oswal points out in his blog. So expect a continued reliance on Wi-Fi, at least for the time being. Eventually, though, once 5G gets up to speed (literally and figuratively), Oswal believes we're in for a "period of innovation in ...

One of the most attractive benefits to intent-based networking (IBN) is its ability to relieve the IT department of updating all networking devices to complete an over-arching business objective. IBN is defined as a self-driving network that automatically applies business intent to various network devices across a network without having to rely on command-line interface (CLI). Instead, the network administrator writes intent in plain language or via a graphical interface. In simple terms, IBN is about telling the software what the business intent is with the software automatically applying the intent across the network. IBN is achievable through orchestration and machine learning (ML).

In my opening blog post this year I wrote about the many issues the market seemed to indicate would be important in 2018. Obviously, I mentioned SD-WAN technology, 5G and Big Data, and I also talked about Artificial Intelligence (AI) and a few others. However, there is one area that I did not mention, which is intent-based networking (IBN). Perhaps it may be early to call 2018 the year of intent-based networking, but some important analysts have said that this is one of the next big things to come! So, what exactly is intent-based networking or IBN?

Cisco this week jumped head first into the intent-based networking market, saying the technology that uses machine learning and advanced automation to control networks could be a major shift in how networks are managed. But what exactly is intent-based networking? MORE AT NETWORK WORLD: Why Cisco's new intent-based networking could be a big deal Gartner Research Vice President Andrew Lerner says intent-based networking systems (IBNS) are not new, and in fact the ideas behind IBNS have been around for years. What's new is that machine learning algorithms have advanced to a point where IBNS could become a reality soon. Fundamentally, an IBNS is the idea of a network administrator defining a desired state of the network, and having automated network orchestration software implement those policies.

Robustness is a commonly bruited property of neural networks; in particular, a folk theorem in neural computation asserts that neural networks-in contexts with large interconnectivity-continue to function efficiently, albeit with some degradation, in the presence of component damage or loss. A second folk theorem in such contexts asserts that dense interconnectivity between neural elements is a sine qua non for the efficient usage of resources. These premises are formally examined in this communication in a setting that invokes the notion of the "devil"