Explore the potential of deep learning in time series analysis through this 48-minute conference talk from MLCon. Discover how neural network architectures can revolutionize the processing of sequential data, including time series and digital signals. Learn about the main sources of time series data, basic algorithms, and how deep learning can improve upon traditional methods. Examine various tasks in time series analysis, such as classification, prediction, anomaly detection, and simulation, and understand how deep learning techniques can achieve state-of-the-art results. Gain insights into applications across different domains, including brain activity analysis, ECG interpretation, and wound assessment. Delve into the challenges and opportunities of applying deep learning to time series data, with speaker Oleksandr Honchar sharing both successful use cases and failed attempts. No prior experience in time series or signal processing is required to benefit from this comprehensive overview of deep learning's impact on sequential data analysis. Read more