{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Graded Exercise 3: Perceptual Decision Making"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "first name: ...\n",
    "\n",
    "last name: ...\n",
    "\n",
    "sciper: ...\n",
    "\n",
    "date: ...\n",
    "\n",
    "*Your teammate*\n",
    "\n",
    "first name of your teammate: ...\n",
    "\n",
    "last name of your teammate: ...\n",
    "\n",
    "sciper of your teammate: ...\n",
    "\n",
    "\n",
    "** Remember **\n",
    "\n",
    "If you are asked for plots: The appearance of the plots (labelled axes, ** useful scaling **, etc.) is important!\n",
    "\n",
    "If you are asked for discussions: Answer in a precise way and try to be concise. \n",
    "\n",
    "\n",
    "** Submission **\n",
    "\n",
    "Rename this notebook to Ex3_FirstName_LastName_Sciper.ipynb and upload that single file on moodle before the deadline.\n",
    "\n",
    "Exercise instructions are given in this notebook file.\n",
    "\n",
    "** Rules: **\n",
    "\n",
    "1) You are strongly encouraged to work in groups of 2. You are allowed to work alone. Groups of 3 or more are NOT allowed\n",
    "\n",
    "2) If you work in a group of 2, BOTH people should upload the same notebook file  \n",
    "\n",
    "3) If you work alone, you can't share your notebook file with anyone else\n",
    "\n",
    "4) Discussion between groups is encouraged, but you can't share your code or text\n",
    "\n",
    "5) The points assigned to each exercise are indicated in the notebook file\n",
    "\n",
    "6) You should upload a jupyter notebook file with all code run and picture visible. We are not going to run your notebook.\n",
    "\n",
    "7) Read carefully the instructions at the beginning of the notebook file, answer in a clear and concise way to open questions\n",
    "\n",
    "8) You have to understand every line of code you write in this notebook. We will ask you questions about your submission during a fraud detection session during the last week of the semester\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## References\n",
    "[1] Wong, K.-F. & Wang, X.-J. A Recurrent Network Mechanism of Time Integration in Perceptual Decisions. J. Neurosci. 26, 1314–1328 (2006).\n",
    "\n",
    "[2] Parts of this exercise and parts of the implementation are inspired by material from *Stanford University, BIOE 332: Large-Scale Neural Modeling, Kwabena Boahen & Tatiana Engel, 2013, online available.*"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 13. Perceptual Decision Making (Wong & Wang)\n",
    "\n",
    "In this exercise we study decision making in a network of competing populations of spiking neurons. The network has been proposed by Wong and Wang in 2006 [1] as a model of decision making in a visual motion detection task. The decision making task and the network are described in the [book](http://neuronaldynamics.epfl.ch/online/Ch16.html) and in the original publication (see References [1]).\n",
    "\n",
    "\n",
    "![test](https://neuronaldynamics-exercises.readthedocs.io/en/latest/_images/DecisionMaking_PhasePlane_3.png)\n",
    "*Decision Space. Each point represents the firing rates of the two subpopulations “Left” and “Right” at a given point in time (averaged over a short time window). The color encodes time. In this example, the decision “Right” is made after about 900 milliseconds.*\n",
    "\n",
    "\n",
    "The parameters of our implementation differ from the original paper. In particular, the default network simulates only 480 spiking neurons which leads to relatively short simulation time even on less powerful computers.\n",
    "\n",
    "**Book chapters**\n",
    "\n",
    "Read the introduction of [chapter 16](https://neuronaldynamics.epfl.ch/online/Ch16.html), Competing populations and decision making. To understand the mechanism of decision making in a network, read 16.2, Competition through common inhibition. You may also want to read the original publication, References [1].\n",
    "\n",
    "**Python classes**\n",
    "\n",
    "The module [`competing_populations.decision_making`](https://neuronaldynamics-exercises.readthedocs.io/en/latest/modules/neurodynex.competing_populations.html#neurodynex.competing_populations.decision_making.sim_decision_making_network) implements the network adapted from References [1,2]. \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "stimulus start: 0.15, stimulus end: 0.5\n",
      "simulating 426 neurons. Start: Fri Apr 12 10:11:45 2019\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO       width adjusted from 20. ms to 20.1 ms [brian2.monitors.ratemonitor.adjusted_width]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "sim end: Fri Apr 12 10:12:26 2019\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x288 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x288 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "import brian2 as b2\n",
    "from neurodynex.tools import plot_tools\n",
    "from neurodynex.competing_populations import decision_making\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "\n",
    "decision_making.getting_started()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Before we can analyse the decision making process and the simulation results, we first need to understand the structure of the network and how we can access the state variables of the respective subpopulations.\n",
    "\n",
    "<img src=\"https://neuronaldynamics-exercises.readthedocs.io/en/latest/_images/DecisionMaking_NetworkStructureAll.png\" alt=\"drawing\" width=\"500\"/>\n",
    "*Network structure. The excitatory population is divided into three subpopulations, shown in the next figure.*\n",
    "\n",
    "<img src=\"https://neuronaldynamics-exercises.readthedocs.io/en/latest/_images/DecisionMaking_NetworkStructureDetail.png\" alt=\"drawing\" width=\"500\"/>\n",
    "\n",
    "*Structure within the excitatory population. The “Left” and “Right” subpopulations have strong recurrent weights ($w^+>w^0$) and weak projections to the other ($w^−<w^0$). All neurons receive a poisson input from an external source. Additionally, the neurons in the “Left” subpopulation receive poisson input with some rate $\\nu_{left}$; the “Right” subpopulation receives a poisson input with a different rate $\\nu_{right}$*"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3.1.1. Question: Understanding Brian2 Monitors"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "The network shown in the figure above is implemented in Brian2 in the function `competing_populations.decision_making.sim_decision_making_network()`. Each subpopulation is a Brian2 NeuronGroup. Look at the source code of the function `sim_decision_making_network()` to answer the following questions:\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**[1+1+1]**\n",
    "\n",
    "- For each of the four subpopulations, find the variable name of the corresponding NeuronGroup.\n",
    "- Each NeuronGroup is monitored with a PopulationRateMonitor, a SpikeMonitor, and a StateMonitor. Find the variable names for those monitors. Have a look at the [Brian2](https://brian2.readthedocs.io/en/stable/user/recording.html) documentation if you are not familiar with the concept of monitors.\n",
    "- Which state variable of the neurons is recorded by the StateMonitor?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Answer**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**[2+1]**\n",
    "\n",
    "\n",
    "- Setting `N_excit=380`how many neurons would be in each excitatory subpopulations?\n",
    "- To keep the default balance between inhibition and excitation what would be the number of inhibitory neurons?\n",
    "\n",
    "**Answer**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For the rest of the miniproject, use `N_Excit=380`, and the found value `N_Inhib`."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "### 3.1.2\\. Question: Accessing a dictionary to plot the population rates[¶](#question-accessing-a-dictionary-to-plot-the-population-rates \"Permalink to this headline\")\n",
    "\n",
    "The monitors are returned in a [Python dictionary](https://docs.python.org/3/tutorial/datastructures.html?highlight=dictionary#dictionaries) providing access to objects by name. Read the [Python documentation](https://docs.python.org/3/tutorial/datastructures.html?highlight=dictionary#dictionaries) and look at the code block below or the function [`competing_populations.decision_making.getting_started()`](https://neuronaldynamics-exercises.readthedocs.io/en/latest/modules/neurodynex.competing_populations.html#neurodynex.competing_populations.decision_making.sim_decision_making_network) to learn how dictionaries are used.\n",
    "\n",
    "**[6]**\n",
    "\n",
    "*   Extend the following code block to include plots for all four subpopulations.\n",
    "*   Run the simulation for 800ms. What are the “typical” population rates of the four populations towards the end of the simulation? (In case the network did not decide, run the simulation again).\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "results = decision_making.sim_decision_making_network(N_Excit=???,N_Inhib=??? ,t_stimulus_start= 50. * b2.ms,\n",
    "                                                      coherence_level=0.6, max_sim_time=???)\n",
    "plot_tools.plot_network_activity(???, ???,\n",
    "                                 ???, t_min=0. * b2.ms, avg_window_width=2. * b2.ms,\n",
    "                                 sup_title=???)\n",
    "\n",
    "\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Answer**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**[3+2+1]**\n",
    "\n",
    "*   Without running the simulation again, but by using the same `results` [dictionary](https://docs.python.org/3/tutorial/datastructures.html?highlight=dictionary#dictionaries), plot the 4 population rates on the same graph using 5 different values of `avg_window_width`. \n",
    "*   Interpret the effect of a very short and a very long averaging window.\n",
    "*   Find a value `avg_window_width` for which the population activity plot the population activity plot gives meaningful rates.\n",
    "\n",
    "Remark: you should use the function [`PopulationRateMonitor.smooth_rate()`](https://brian2.readthedocs.io/en/2.0.1/user/recording.html#recording-population-rates). \n",
    "\n",
    "`avg_window_width = 123*b2.ms`\n",
    "\n",
    "`sr = results[\"rate_monitor_A\"].smooth_rate(window=\"flat\", width=avg_window_width)/b2.Hz`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Answer**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "## 3.2. Exercise: Stimulating the decision making circuit\n",
    "\n",
    "The input stimulus is implemented by two inhomogenous Poisson processes: The subpopulation “Left” and “Right” receive input from two different PoissonGroups (see Figure “Network Structure”). The input has a `coherence level c` and is noisy. We have implemented this in the following way: every 30ms, the firing rates $\\nu_{left}$ and  $\\nu_{right}$  of each of the two PoissonGroups are drawn from a normal distribution:\n",
    "\n",
    "\n",
    "$\\nu_{left}\\sim\\mathcal{N}(\\mu_{left},\\sigma^2)$\n",
    "\n",
    "$\\nu_{right}\\sim\\mathcal{N}(\\mu_{right},\\sigma^2)$\n",
    "\n",
    "$\\mu_{left}=\\mu_0*(0.5+0.5c)$\n",
    "\n",
    "$\\mu_{right}=\\mu_0*(0.5-0.5c)$\n",
    "\n",
    "$c\\in[-1,+1]$\n",
    "\n",
    "The coherence level $c$, the maximum mean $\\mu_0$ and the standard deviation $\\sigma$ are parameters of `sim_decision_making_network()`.\n",
    "\n",
    "## 3.2.1. Question: Coherence Level\n",
    "\n",
    "**[1+1]**\n",
    "\n",
    "- From the equation above, express the difference $\\mu_{left}−\\mu_{right}$ in terms of $\\mu_0$ and $c$.\n",
    "\n",
    "- Find the distribution of the difference $\\nu_{left}−\\nu_{right}$. Hint: the difference of two Gaussian distributions is another Gaussian distribution.\n",
    "\n",
    "\n",
    "Now look at the documentation of the function [`sim_decision_making_network()`](https://neuronaldynamics-exercises.readthedocs.io/en/latest/modules/neurodynex.competing_populations.html#neurodynex.competing_populations.decision_making.sim_decision_making_network) and find the default values of μ0 and σ. Using those values, answer the following questions:\n",
    "\n",
    "**[1+1]**\n",
    "\n",
    "- What are the mean firing rates (in Hz) $\\mu_{left}$ and $\\mu_{right}$ for the coherence level c= +0.2?\n",
    "- For c= +0.2, how does the difference $\\mu_{left}$ and $\\mu_{right}$ compare to the variance of $\\nu_{left}$ and $\\nu_{right}$.\n",
    "\n",
    "**Answer**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3.2.2. Question: Input stimuli with different coherence levels\n",
    "\n",
    "**[2+1+1]**\n",
    "\n",
    "Run a few simulations with `c=-0.6` and `c=+0.1`. Plot the network activity.\n",
    "\n",
    "- Does the network always make the correct decision?\n",
    "- Look at the population rates and estimate roughly how long it takes the network to make a decision.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Answer**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3.3. Exercise: Decision Space\n",
    "\n",
    "We can visualize the dynamics of the decision making process by plotting the activities of the two subpopulations “Left” / “Right” in a phase plane (see figure at the top of this page). Such a phase plane of competing states is also known as the _Decision Space_. A discussion of the decision making process in the decision space is out of the scope of this exercise but we refer to Reference [1]\n",
    "\n",
    "\n",
    "\n",
    "### 3.3.1 Question: Plotting the Decision Space\n",
    "\n",
    "**[5+2]**\n",
    "\n",
    "*   Write a function that takes two [RateMonitors](http://brian2.readthedocs.io/en/2.0.1/user/recording.html#recording-population-rates) and plots the _Decision Space_.\n",
    "*   Add a parameter `avg_window_width` to your function (same semantics as in the exercise above). Run a few simulations and plot the phase plane for different values of `avg_window_width`. Use `c=0.3`.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**[2]**\n",
    "\n",
    "*   We can use a rate threshold as a decision criterion: We say the network has made a decision if one of the (smoothed) rates crosses a threshold. What are appropriate values for `avg_window_width` and `rate` to detect a decision from the two rates?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3.3.2 Question: Implementing a decision criterion\n",
    "\n",
    "**[4]**\n",
    "\n",
    "*   Using your insights from the previous questions, implement a function **get_decision_time** that takes two [RateMonitors](http://brian2.readthedocs.io/en/2.0.1/user/recording.html#recording-population-rates) , a `avg_window_width` and a `rate_threshold`. The function should return a tuple (decision_time_left, decision_time_right). The decision time is the time index when some decision boundary is crossed. Possible return values are (1234.5ms, 0ms) for decision “Left”, (0ms, 987.6ms) for decision “Right” and (0ms, 0ms) for the case when no decision is made within the simulation time. A return value like (123ms, 456ms) is an error and occurs if your function is called with inappropriate values for `avg_window_width` and `rate_threshold` and must be considered as such in your code. \n",
    "\n",
    "The following code fragments could be useful:\n",
    "\n",
    "`smoothed_rates_A = rate_monitor_A.smooth_rate(window=\"flat\", width=avg_window_width) / b2.Hz\n",
    "idx_A = numpy.argmax(smoothed_rates_A > rate_threshold/b2.Hz)\n",
    "t_A = idx_A * b2.defaultclock.dt`\n",
    "\n",
    "Run a few simulations to test your function."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**[1]**\n",
    "\n",
    "How do the percentage of correct answer evolve with respect to `c`.\n",
    "\n",
    "**Answer**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 3.5 Winner-take-all in artificial Neural network\n",
    "\n",
    "Winner-take-all is a computational models of neural networks by which neurons in a layer compete with each other for activation.  \n",
    "\n",
    "In the classical form, only the neuron with the highest activation stays active while all other neurons are shut down; however, other variations allow more than one neuron to be active. It may be link to a decision making task as the decision is defined by the winner neuron.\n",
    "\n",
    "Consider a network of formal neurons described by activities\n",
    "\\begin{equation}\n",
    "  A_i= g(h_i) = \\left\\{\n",
    "    \\begin{array}{ll}\n",
    "    0 & \\mbox{ for } h_i<1 \\\\\n",
    "    h_i -1& \\mbox{ for } 1<h_i<2\\\\ \n",
    "    1& \\mbox{ for } h_i>2\\\\ \n",
    "    \\end{array}\n",
    "  \\right. .\n",
    "\\end{equation}\n",
    "for $i=1,...,N$, where $N$ is the number of neurons.\n",
    "The input potential $h_i$ evolves with the following update rule: \n",
    "\n",
    "$\\tau\\frac{dh_i}{dt}= f(h) = -h_i + w_0g(h_i) - \\alpha\\sum_{j \\neq i} g(h_j) + h_{ext,i}$. (1)\n",
    "\n",
    "where $w_0$ and $\\alpha$ are parameters greater than 0. For $t \\leq 0$ the external input vanishes and all the neurons are at their resting potential $h_i=0$. For $t > 0$ we have $h_{ext,i} = 1 + (0.5)^i$\n",
    "\n",
    "**[1+1]**\n",
    "\n",
    "- what is the intrepretation of $w_0$ and $\\alpha$\n",
    "\n",
    "**Answer**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3.5.1 Implementation of simple winner-take-all\n",
    "\n",
    "**[2+3]**\n",
    "\n",
    "- Implement the function $g(h_i)$ which takes in argument $h_i$. \n",
    "\n",
    "- Implement the update rule in a simple Euler fashion i.e. $h_i(t+\\Delta t) = h_i(t) + \\eta f(h(t))$. \n",
    "\n",
    "We impose here $\\eta = \\Delta t / \\tau = 0.01$.\n",
    "\n",
    "**Answer**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def g(???): ## gain function ##\n",
    "    return  ???  ## takes a vector returns a vector of all neurons ##\n",
    "\n",
    "def update_rule(???): ## update rule for the potentials ##\n",
    "    return ??? ## takes a vector returns a vector of all neurons ##"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**[3]**\n",
    "- Now run a simulation for a small network of `10 neurons` for $500$ time steps, with $w_0=2$ and $\\alpha=1$. and plot the resulting activities. Comment on what you observe.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "N_neuron = ???\n",
    "alpha = ???\n",
    "w0 = ???\n",
    "eta = ???\n",
    "h = ??? ## initial potential of the neurons (vector) ##\n",
    "dic_evolution = {}\n",
    "dic_evolution['activities'] = []\n",
    "for i in range(???):\n",
    "    \n",
    "    ## Implement the update ##\n",
    "    \n",
    "    dic_evolution['activities'] += [???] ## add the activity at time t to the list ##"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Answer**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3.5.2 Influence of the parameters\n",
    "\n",
    "**[1+1]**\n",
    "\n",
    "- What happens if you decrease $\\alpha$? (Keep $w_0=2$)\n",
    "\n",
    "- What happens if you decrease $w_0$? (Keep $\\alpha=1$)\n",
    "\n",
    "justify your answer by the means of plots.\n",
    "\n",
    "**Answer**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**[2+2]**\n",
    "\n",
    "Derive sufficient conditions so that the only fixed point is $A_i = \\delta_{i,1}$, i.e. only the neuron with the highest external input can be a winner. Write the derived answer and show the critical behaviour.\n",
    "\n",
    "hints:\n",
    "\n",
    "- only $h_1$ can be greater than 2 (i.e. $A_1 =1$) all the other must have zero acitivity\n",
    "\n",
    "**Answer**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3.5.3 Implementation of a K-winners-take-all\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**[2+1+1+2]**\n",
    "\n",
    "By adapting both $w_0$ and $\\alpha$ one can update our previous network to obtain a K-winners-take-all neuron, i.e. K neurons will reach an activity of 1.\n",
    "\n",
    "- Find a necessary condition on $w_0$ as a function of $\\alpha$ such that exactly $K$ neurons have activity 1, where the others have all 0 activity.\n",
    "\n",
    "hint:\n",
    "- the fixed point of $h_K$ has to be greater than 2 (i.e. $A_K =1$), the K-1 neurons which recive a stronger input have also activity 1 and all the other must have zero acitivity.\n",
    "\n",
    "\n",
    "- One can persuade himself that the following equations are sufficient conditions for alpha. Solve them then **implement the network for K=3 winners**.\n",
    "\\begin{align}\n",
    "-K\\alpha+1+0.5^{K+1} \\leq 1 \\\\\n",
    "-(K-1)\\alpha+1+0.5^K>1 \\\\\n",
    "\\end{align}\n",
    "- plot your results, showing the critical behaviour.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3.5.4 stochastic input, winners-take-all\n",
    "\n",
    "In biological neural network we often consider noisy input coming from background activity. In this exercice we want you to observe the influence of the noise in the winner-take-all process. \n",
    "\n",
    "To do so consider $w0=2$ and $\\alpha=2$ and define the external input as $h_{ext,k}= 1 + 0.5^k +\\xi(t)$, where $\\xi(t)$  is a white noise of standard deviation 1.\n",
    "\n",
    "**[2+1]**\n",
    "\n",
    "- implement the white noise and run a simulation, what do you observe ? Compute the probability of the neuron 1 to win\n",
    "\n",
    "**Answer**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**[3]**\n",
    "\n",
    "- Now consider the external input $h_{ext,k}= 1 + 0.9^k +\\xi(t)$. Run a few simulations to estimate the probability of neuron 1 to win. Comment on the differences you observe [max 5 lines]\n",
    "\n",
    "**Answer**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": []
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}