sushirain

5 ms·

1.
by sushirain 8y ago
The bot can gather this restaurant specific information over several conversations with the restaurant. This wasn't possible before. This domain isn't too wide.
2.
by sushirain 9y ago
You can use a character RNN that reads the fragmented input char by char, and feed its output to your NN.
3.
by sushirain 9y ago
Is it possible to create an adversarial example without access to the weights of a model, and without being able to forward many images through it?
4.
Autonomous Vehicle Implementation Predictions [pdf](vtpi.org)
2 points by sushirain 9y ago|0 comments
5.
by sushirain 9y ago
Can autonomous vehicles work without deep reinforcement learning? I thought that things like negotiating entry into a crossroad required DRL.
6.
by sushirain 9y ago
Thanks. I found that when they transitioned to Hardware2/AP2 they cut out most autopilot functionality, and since then have been gradually releasing updates that bring back that functionality.
7.
by sushirain 9y ago
I couldn't find a reference to this:"Tesla no longer sells cars with automatic driving capability." Can you point to one?
8.
The Global Risks Landscape 2017(reports.weforum.org)
1 points by sushirain 9y ago|0 comments
9.
by sushirain 9y ago
Very interesting. I wonder if they tried to predict part-of-speech tags.
10.
by sushirain 9y ago
The article says 5 years old, so it's probably not deep learning.
11.
by sushirain 10y ago
This is the relevant chart: http://www.jcmit.com/mem2015.htm DIMM seems to go down slower lately.
12.
by sushirain 10y ago
Would Apple?
13.
by sushirain 10y ago
My summary of the slides: to achieve "common sense", Unsupervised Learning is the the next step. Use all information in the data for predicting everything you can including the past. Use adversarial networks to do that.
14.
by sushirain 10y ago
If you don't have financial obligations, you could quit. Otherwise, maybe it's better to join an open source project on the side for a year. It takes years to go from noob to proficiency in coding, machine learning, cryptography.
15.
by sushirain 10y ago
Deep learning is already being monetized, e.g., at Google.
16.
by sushirain 10y ago
Vehicle detection: detect any vehicle, even from the side, even if its shape is rare, even at dark, etc. This could be solved with data. Control: when to yield, without watching the face of the other driver, etc. Edge cases: obeying a polic
17.
by sushirain 10y ago
Weather conditions are a sensor / computer vision /control problem. It is far from solved.
18.
by sushirain 10y ago
The difference between 98% and 99.999% is very difficult to solve, and it's not going to happen in the next 5 years. LIDAR can't help, for example, with obeying a police officer gesture.
19.
'Big Bang Theory' Sheldon Prequel Series in the Works at CBS(hollywoodreporter.com)
7 points by sushirain 10y ago|6 comments
20.
SoftBank and Saudi Arabia plan $100bn tech fund(ft.com)
3 points by sushirain 10y ago|1 comments
21.
by sushirain 10y ago
> Son said that he teamed up with a Saudi Arabian group to raise a $100 billion investment fund as part of a plan to move to the next stage. Also see: "SoftBank and Saudi Arabia plan $100bn tech fund" https://www.ft.
22.
SoftBank’s Masayoshi Son is investing to make the ‘Singularity’ happen(venturebeat.com)
3 points by sushirain 10y ago|1 comments
23.
by sushirain 10y ago
Why not use Medium instead of twitter? "Marcin Wichary, the lead typographer and designer at Medium, is obsessed with all things typeface, including typewriters. So when he chanced upon Museu de la Tècnica in a small Spanish town, noth
24.
by sushirain 10y ago
Can you be more specific?
25.
Ask HN: How often women read HN? Participate?
21 points by sushirain 10y ago|24 comments
26.
by sushirain 10y ago
The linguistic patterns that you mentioned (parts of speech, etc.) are too simplistic to describe language's idiosyncrasy and they ignore language acquisition. Machine learning and data-driven methods require models with millions of pa
27.
by sushirain 10y ago
Reminds me of Turing's morphogenesis. https://www.wired.com/2011/02/turing-patterns/
28.
by sushirain 10y ago
For example, ResNet from 2015 had 152 layers. A real neuron takes in the order of 10ms to integrate and fire to the next neuron. Many subconscious reactions take less than 1 sec, which leaves time to a chain of length less than 100. Note th
29.
by sushirain 10y ago
Your arguments are compelling. I hope others will embrace them more.
30.
OpenAI Requests for research(openai.com)
3 points by sushirain 10y ago|0 comments

More ›