• Continued with explorations of dialogue related problems that can potentially turn out to be hard or interesting. A few updates are here. Works in dialogue alignment have nice threads to follow.

• Probing neural network comprehension of natural language arguments (niven2019probing).
• Alignment in dialogue (garrod2007alignment).
• Statistical user simulation with a hidden agenda (schatzmann2007statistical).
• Collaborative Multi-Agent Dialogue Model Training Via Reinforcement Learning (papangelis2019collaborative).

• Finished porting my player to lisp completely. Onto more important things now.

• Refusal of Work: Free Time and the Pressures of Employability. Among other things, I got a nice take on education's purpose here, which is to enable enjoyment of life's more complex pleasures.
• What can Schrödinger’s cat say about 3D printers on Mars? Not a lot of new information but I like the way of differentiating the words life (lineage of functional information) and alive (things that maintain themselves).

Bibliography

• [niven2019probing] @miscniven2019probing, Author = Niven, Timothy and Kao, Hung-Yu, Title = Probing Neural Network Comprehension of Natural Language Arguments, Year = 2019, Eprint = arXiv:1907.07355,
• [garrod2007alignment] Garrod & Pickering. 2007. "Alignment in dialogue." The Oxford handbook of psycholinguistics, , 443-451. link. doi.
• [schatzmann2007statistical] Schatzmann, Thomson & Young. 2007. "Statistical user simulation with a hidden agenda." Proc SIGDial, Antwerp, 273282(9), link. doi.
• [papangelis2019collaborative] @miscpapangelis2019collaborative, Author = Papangelis, Alexandros and Wang, Yi-Chia and Molino, Piero and Tur, Gokhan, Title = Collaborative Multi-Agent Dialogue Model Training Via Reinforcement Learning, Year = 2019, Eprint = arXiv:1907.05507,