Entries by Falk Pollok

PalmQA, a Question-Answering Ensemble for e-Learning and Research

My first bigger AI system developed under the supervision of Mohamed Amine Chatti & Ulrik Schroeder. “The scientific mind does not so much provide the right answersas ask the right questions.” Claude Lévi-Strauss 1  Introduction From Emanuel Goldberg’s Statistical Machine and Vannevar Bush’s Memex over collaboration systems like Ted Nelson’s Project Xanadu, Douglas Engelbart’s oN-Line […]

EXSUM: Audio Support in Containers

Not many people seem to know how to support audio in containers. Here is an example for forwarding Pulseaudio into a container. This will open a shell into the container. However, before going there run the following on host (e.g. in new shell): Now play audio in container, e.g. online radio via MPlayer:

EXSUM: Ray on Minikube

I was asked about how to setup Kubernetes and deploy Ray on it. The last time I did this was before this was officially supported – I just manually built a container with Ray and deployed it as master on K8s master and workers on K8s workers. However, over the last 3 years or so […]

Side Track: Augmenting Text Adventures with Illustrative Art

Ryan Murdoch recently published another one of his cool projects in which he uses OpenAI’s CLIP to steer BigGAN towards matching textual inputs. The outputs have a surreal-expressionist quality to them which makes them interesting to look at. So the idea struck me to use them for text adventure illustration. I have taken Zork (which […]

Reinforcement Learning Frameworks – An Overview

Abstract Reinforcement Learning (RL) has seen renewed interest sparked by the successful combination of RL with neural models as well as Monte-Carlo Tree Search (MCTS). At first, this development was largely restricted to playing traditional games and video games, but successively one can observe more widespread usage in industry as well from robotics and autonomous cars to datacenter and […]

Kubernetes Storage Considerations for AI Workloads

This is a repost of my 3-part blog post from November 2018. We discuss storage options for AI and particularly deep learning and discuss how to avoid a few common pitfalls, esp. regarding small file support. The structure is as follows: Fundamentals of IBM Cloud Storage Solutions for Kubernetes including Cloud Object, File and Block […]