numpde

Reinforcement learning through imitation of successful peers

Introduction

Reinforcement learning is a mode of machine learning driven by the feedback from the environment on how good a string of actions of the learning agent turns out to be.

We consider here a reinforcement learning mechanism for neural networks

numpde

#!/usr/bin/python3

# CC-BY-4.0

import sys, argparse
from random import shuffle, randrange, choice

# Collate lines, separating them by a space
s = ' '.join(sys.stdin.readlines())
# Remove non-text

numpde

#!/usr/bin/python3

# Histograms for the number of types V(N) within the first N tokens.
# Comparison of natives vs learners.

# R. Andreev, 2017-05-11 (first version), CC BY 4.0

# Designed for the ANGLISH corpus [Tortel 2008, via N. Ballier & P. Lisson]
# The texts are expected to be located in ./ANGLISH/*.txt

numpde

#/bin/bash

# License-free

# Example: addpath PATH "/usr/bin"
addpath() {
        varname=$1
        export $1="$2:${!varname}"
}

numpde

function h = renice(h)
% function h = renice(h)
%   
%   h is a handle returned by PLOT and such
%
%   RA, Apr 2008 -- Aug 2016
%
%   License: CC-BY-4.0
    
    set(0, 'DefaultTextInterpreter', 'tex');

numpde

def betti(G, C=None, verbose=False):
    # G is a networkx graph
    # C is networkx.find_cliques(G)

    # RA, 2017-11-03, CC-BY-4.0

    # Ref: 
    # A. Zomorodian, Computational topology (Notes), 2009
    # http://www.ams.org/meetings/short-courses/zomorodian-notes.pdf

numpde

# Find the rank of a binary matrix over Z/2Z
# (conceptual implementation)
#
# RA, 2017-11-07 (CC-BY-4.0)
#
# Adapted from
#   https://triangleinequality.wordpress.com/2014/01/23/computing-homology/
#
def binary_rank(M) :
    

numpde

# Compute the Betti numbers of a graph over Z/2Z.
# 
# If G is a networkx graph then set
#   C = networkx.find_cliques(G)
# This enumerates maximal cliques.
#
# Pass C to the function betti_bin,
# which returns a list of Betti numbers.
#
# RA, 2017-11-08 (CC-BY-4.0)

numpde

#!/bin/bash

# Suppose you wish to connect to a server X on port 23
# but you have to do this via another server Y like this:
#
#   localhost> ssh -p 22 username@Y
#           Y> ssh -p 23 username@X
#
# Instead, you can create an ssh tunnel on a local port, say 9999:
#   localhost> ssh -nNT -L 9999:X:23 -p 22 username@Y

numpde

#!/usr/bin/python3

# AUTHOR, DATE

## ================== IMPORTS :

pass
import inspect