However, we have identified that the effectiveness of these rewards declines as the environmental complexity rises. Therefore, we present a novel USD algorithm, skill disco very with gui dance ( DISCO-DANCE), which (1) selects the guide skill that possesses the highest potential to reach unexplored states, (2) guides other skills to follow guide skill, then (3) the guided skills are dispersed to maximize their discriminability in unexplored states. Empirical evaluation demonstrates that DISCO-DANCE outperforms other USD baselines in challenging environments, including two navigation benchmarks and a continuous control benchmark.

Interactive programming environments are powerful tools for promoting innovative network thinking, teaching science of complexity, and exploring emergent phenomena. This paper reports on our recent development of the deterministic random walk model in NetLogo, a leading platform for computational thinking, eco-system thinking, and multi-agent cross-platform programming environment. The deterministic random walk is foundational to modeling dynamical processes on complex networks. Inspired by the temporal visualizations offered in NetLogo, we investigated the relationship between network topology and diffusion saturation time for the deterministic random walk model. Our analysis uncovers that in Erd\H{o}s-R\'{e}nyi graphs, the saturation time exhibits an asymmetric pattern with a considerable probability of occurrence. This behavior occurs when the hubs, defined as nodes with relatively higher number of connections, emerge in Erd\H{o}s-R\'{e}nyi graphs. Yet, our analysis yields that the hubs in Barab\'{a}si-Albert model stabilize the the convergence time of the deterministic random walk model. These findings strongly suggest that depending on the dynamical process running on complex networks, complementing characteristics other than the degree need to be taken into account for considering a node as a hub. We have made our development open-source, available to the public at no cost at https://github.com/bravandi/NetLogo-Dynamical-Processes.

Ontology matching is the process of finding semantic correspondences between entities from different ontologies. As an effective solution to linking different heterogeneous ontologies, ontology matching has attracted considerable attentions in recent years. In this paper, we propose a novel graph-based approach to ontology matching problem. Different from previous work, we formulate ontology matching as a random walk process on the association graph constructed from the to-be-matched ontologies. In particular, two variants of the conventional random walk process, namely, Affinity-Preserving Random Walk (APRW) and Mapping-Oriented Random Walk (MORW), have been proposed to alleviate the adverse effect of the false-mapping nodes in the association graph and to incorporate the 1-to-1 matching constraints presumed in ontology matching, respectively. Experiments on the Ontology Alignment Evaluation Initiative (OAEI) datasets show that our approach achieves a competitive performance when compared with state-of-the-art systems, even though our approach does not utilize any external resources.

In my other posts, I have covered topics such as: How to combine machine learning and physics, and how machine learning can be used for production optimization as well as anomaly detection and condition monitoring. But in this post, I will discuss some of the common pitfalls of machine learning for time series forecasting. Time series forecasting is an important area of machine learning. It is important because there are so many prediction problems that involve a time component. However, while the time component adds additional information, it also makes time series problems more difficult to handle compared to many other prediction tasks.

There are several types of models that can be used for time-series forecasting. In this specific example, I used a Long short-term memory network, or in short LSTM Network, which is a special kind of neural network that make predictions according to the data of previous times. It is popular for language recognition, time series analysis and much more. However, in my experience, simpler types of models actually provide just as accurate predictions in many cases. Due to their sequential nature, TDNN's are implemented as a feedforward neural network instead of a recurrent neural network.