As machine learning technologies become more prevalent, the risk of attacks continues to rise. Which types of attacks on ML systems exist, how do they work, and which is the most dangerous? ML Conference speaker David Glavas answered our questions.
Take a tour of ycrash in this article by Ram Lakshmanan. ycrash helps capture critical artifacts, including garbage collection logs, thread dumps, core dumps, heap dumps, disk usage, and more when the problem happens. It applies machine learning algorithms and generates a report which gives you a complete view of the problem, down to the lines of code that caused it.
Can AI play and complete a game? Juantomás Garcia Molina’s session from the Machine Learning Conference looks at developing artificial intelligence that can complete the first 3-D RPG, created in 1987. Many people had difficulty completing this technological wonder, so how will artificial intelligence fare?
Machine learning can help predict things dependent on time such as taxi demand. Time series forecasting has always been an important field in machine learning and statistics, as it helps us to make decisions about the future. A special field is spatio-temporal forecasting, where predictions are not only made on the temporal dimension, but also on a regional dimension.
In this session, speaker Michael Kieweg will discuss data and AI and the relationship between the two. Get comfortable and watch his session from the Machine Learning conference where he discusses how to tackle challenges related to data quality and how to use data for better artificial intelligence performance.
Increasingly large and diverse data sets allow us to form complex insights. With all this data, why would we limit ourselves by using data sampling instead? Sampling only works when it is put in the hands of data science specialists. In this article, learn about some of the downsides of using data sampling and how it limits and undermines business decisions. Read part one of the case against data sampling.
The full version of GPT-2 is now publicly available, following nearly nine months of heated debates and some smaller model releases. The large-scale unsupervised language model was kept under lock and key for this long as it was deemed too dangerous—a controversial decision that led to backlash from the open source community.
The early bird special for the Machine Learning Conference and the Voice Conference ends on November 7, 2019. This special offer includes a ticket discount of up to €210. Check out the conference programs, the tracks, speakers, and what to expect at these two conferences. A ticket for one includes admission to both!
The newest update for PyTorch-NLP is here. The 0.5.0 update adds support for Python 3.5, PyTorch 1.2, rewrites the README to help new users build an NLP pipeline, and adds some new features. See how PyTorch-NLP helps with natural language processing and how PyTorch compares to similar machine learning frameworks such as TensorFlow.
Google introduced TensorFlow Enterprise, a new collection of machine learning services and products. The beta version includes managed services, and some versions will receive long-term version support for up to three years. At the same time, Google unveiled the new website TensorBoard.dev.
We love open source! And now, two new recently open sourced pieces of software from Netflix have arrived. Mantis allows you to build realtime stream processing applications and Polynote is an IDE-inspired polyglot notebook. See what both of these can do, some of their use cases, and what important features set them apart.
Take a look into the crystal ball. What does Gartner predict for 2020? Here are ten strategic trending technologies that tech leaders should have on their radar in the coming years. Augmented and virtual reality might train employees in the future, voting might become based on a blockchain, and AI security might become the most important factor in risk management.
Voice technologies are often caught between the user’s expectations of a truly personal assistant and the user’s desire for privacy and anonymity. We talked to Jeremy Wilken, moderator of the Design for Voice podcast and speaker at the upcoming VoiceCon Berlin, about a way to reconcile these expectations.
At the moment, big data is very popular and there is a wide variety of products available for handling data. In this article, read a case study about a German startup tackled their data problems and built a common data platform into their architecture. The data platform consists of four components: Ingestion, storage, process, and provisioning.
