dela3499

Research problems

I'd like to understand intelligence. I think there are several areas of research that are especially interesting: conjecture, criticism, realism, problem-solving, and universality.

Conjecture - fast search

We're faced with a constant stream of problems, and we're quite good at finding solutions. How do we quickly and efficiently navigate the space of all possible solutions?

A new problem might be solved with an old solution, or some variant of it, with some ideas added, removed, or tweaked. Analogies or random combinations may help.

dela3499

prompts: 
  - "#business for #"
  - "#tool applied to #"
  - "Modify and resell #product (amount, packaging, marketing)"
  - "Create a shared resource around #"
  - "Bundle # with #"
  - "Unbundle #"
  - "Improve the #quality of #"
  - "Connect people from # and #"
  - "# + #"

dela3499

Our computer programs can't yet match human-level general intelligence, but there are two directions of research I think are especially encouraging.

First, we can extend the capabilities of deep learning systems, which have been wildly successful in many narrow applications. As Francois Chollet argues, deep learning programs are only a subset of all possible programs. In other words, they aren't universal. So, it's not surprising we'd find success in only narrow applications. Humans can presumably learn any program (that can run given the memory and speed limitations of our brains). So, if the geometric transforms of deep learning aren't enough, then where do we go from here?

We already have two examples: RNNs and AlphaGo. Recurrent neural networks are like normal deep learning architectures, except that they effectively contain a for loop. AlphaGo also contains an additional component (tree-search) that isn't purely deep-learned. These examples point the way to finding systems which incorporate ideas fro

dela3499

one but on so a have would would would would day i prioritization are prioritization with not not not not test test test test a problem problem solve solve solve at my think a im do do are about about about about day a a my have would theories theories make like different today today today today not think the or would would would would so are think you my these like like like what a you be be some for want on also one one one one or have would with of also of today some of new can can would i different how there there there you like do have have today not test think think think think would would would would that so have that would new a a a of time time time time have have can test want be theories in in in in in make are to to to there you also my solve want at day in problem are today but one im im im what what not in my what theories theories theories theories how different you day some today and you to have prioritization one new for day today new day not for do day the it time be that problem problem be 

dela3499

from bokeh.plotting import figure, output_file, show, ColumnDataSource
from bokeh.models import HoverTool
from bokeh.io import output_notebook
output_notebook()

source = ColumnDataSource(
        data=dict(
            x=[1, 2, 3, 4, 5],
            y=[2, 5, 8, 2, 7],
            desc=['A', 'b', 'C', 'd', 'E'],

dela3499

var vids = document.getElementsByTagName('video') 

for( var i = 0; i < vids.length; i++ ){ 
    vids.item(i).playbackRate = 2
}

dela3499

dela3499

dela3499

# http://stackoverflow.com/questions/11133290/git-merging-using-meld
[merge]
tool = mymeld
conflictstyle = diff3
[mergetool "mymeld"]
cmd = meld --diff $BASE $LOCAL --diff $BASE $REMOTE --diff $LOCAL $BASE $REMOTE $MERGED

dela3499

import Set exposing (Set)

toggleMember: String -> Set String -> Set String
toggleMember value set = 
  if Set.member value set then
    Set.remove value set
  else
    Set.insert value set