Michael Nielsen

"The first message on the ARPANET was sent by UCLA student programmer Charley Kline, at 10:30 p.m, on October 29, 1969 from Boelter Hall 3420, the school's main building.[6] Supervised by Kleinrock, Kline transmitted from the university's SDS Sigma 7 host computer to the Stanford Research Institute's SDS 940 host computer. The message text was the word "login"; the "l" and the "o" letters were transmitted, but the system then crashed. Hence, the literal first message over the ARPANET was "lo". About an hour later, having recovered from the crash, the SDS Sigma 7 computer effected a full "login". The first permanent ARPANET link was established on November 21, 1969, between the IMP at UCLA and the IMP at the Stanford Research Institute. By December 5, 1969, the entire four-node network was established.[7]" - Michael Nielsen

Atkinson's attitude to the web: "Atkinson feels that if only he'd realized separate cards and stacks could be linked on different people's machines through the Net -- instead of cards and stacks on a particular machine -- he would have created the first Internet browser." - Michael Nielsen

Some evidence to suggest that there's little payoff from larger grants. Lots of caveats, of course. Interesting references, too. - Michael Nielsen

A useful review. - Michael Nielsen

On Google's use of Hinton-Ng-Dean-LeCun's work to do automatic photo tagging. - Michael Nielsen

Reports two experiments in which max pooling handily beat out subsampling. - Michael Nielsen

Agre on what it means to be critical: to be aware of one's own biases, and the cultural assumptions one brings to something. He talks about close reading an AI text, and trying to understand all the sociological background. - Michael Nielsen

On the 2012 Google-Stanford deep learning paper. - Michael Nielsen

More learned features helps a lot; smaller stride length helps a lot. Larger local receptive fields barely helps at all. - Michael Nielsen

Results for MNIST, CIFAR-10, CIFAR-100, STL-10 and SVHN - Michael Nielsen

Simard, Steinkraus, and Platt on the value of image distortion and convolutional networks. - Michael Nielsen