{ "cells": [ { "cell_type": "markdown", "id": "3d531da8-f5e7-4979-9c95-a2dd12c8298e", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "source": [ "(parameters_recovery)=\n", "# Recovering computational parameters from observed behaviours" ] }, { "cell_type": "markdown", "id": "5d880108-f486-4fcc-8bf3-651c6c8b829c", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "source": [ "[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/ilabcode/pyhgf/blob/master/docs/source/notebooks/4-Parameter_recovery.ipynb)" ] }, { "cell_type": "code", "execution_count": 1, "id": "e00c09dd-b0ed-4e32-8bc7-af5868acdcd3", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [ "hide-cell" ] }, "outputs": [], "source": [ "import sys\n", "from IPython.utils import io\n", "if 'google.colab' in sys.modules:\n", "\n", " with io.capture_output() as captured:\n", " ! pip install pyhgf watermark" ] }, { "cell_type": "code", "execution_count": 2, "id": "4bc4fb2a-0224-4374-b046-76ab5904e757", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "An NVIDIA GPU may be present on this machine, but a CUDA-enabled jaxlib is not installed. Falling back to cpu.\n" ] } ], "source": [ "import arviz as az\n", "import jax.numpy as jnp\n", "import matplotlib.pyplot as plt\n", "import numpy as np\n", "import pymc as pm\n", "import seaborn as sns\n", "from pyhgf import load_data\n", "from pyhgf.distribution import HGFDistribution\n", "from pyhgf.math import sigmoid_inverse_temperature\n", "from pyhgf.model import HGF\n", "from pyhgf.response import binary_softmax_inverse_temperature" ] }, { "cell_type": "code", "execution_count": 3, "id": "20aba0ed-e8c6-4276-8778-496528d41232", "metadata": {}, "outputs": [], "source": [ "np.random.seed(123)" ] }, { "cell_type": "markdown", "id": "dbef25ef-44e0-4d35-bfa4-5be3bb3a393e", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "source": [ "An important application of Hierarchical Gaussian Filters consists in the inference of computational parameters from observed behaviours, as well as the inference of data-generating models (e.g. are the participants answering randomly or are they learning environmental volatilities that are better approached with a Rescorla-Wagner or a Hierarchical Gaussian Filter?). **Parameter recovery** refers to the ability to recover true data-generating parameters; **model recovery** refers to the ability to correctly identify the true data-generating model using model comparison techniques. It is often a good idea to test parameter/model recovery of a computational model using simulated data before applying this model to experimental data {cite:p}`RobertCollins2019`. In this tutorial, we demonstrate how to recover some parameters of the generative model of the Hierarchical Gaussian Filter." ] }, { "cell_type": "markdown", "id": "a3a79141-eeda-47fa-bff1-eecd6bf035ee", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "source": [ "## Simulate behaviours from a one-armed bandit task\n", "Using a given task structure, we simulate behaviours from a group of participants assuming that they are updating beliefs of environmental volatility using a two-level Hierarchical Gaussian Filter, using a simple sigmoid as a response function parametrized by an inverse temperature parameter. For each participant, the inverse temperature and the tonic volatility at the second level are free parameters that will be estimated during the inference step." ] }, { "cell_type": "code", "execution_count": 4, "id": "70400fe9-3277-4674-97de-d17543d42054", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "outputs": [], "source": [ "u, _ = load_data(\"binary\") # the vector encoding the presence/absence of association\n", "\n", "N = 20 # the number of agents to simulate\n", "\n", "# sample one value for the inverse temperature\n", "temperatures = np.linspace(0.5, 6.0, num=N)\n", "\n", "# sample one new value of the tonic volatility at the second level\n", "volatilities = np.linspace(-6.0, -1.0, num=N)" ] }, { "cell_type": "code", "execution_count": 5, "id": "7d8a095c-2360-4a12-9539-7a6c94b99c48", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "outputs": [], "source": [ "# create just one default network - we will simply change the values of interest before fitting to save time\n", "agent = HGF(\n", " n_levels=2,\n", " model_type=\"binary\",\n", " initial_mean={\"1\": 0.5, \"2\": 0.0},\n", ")" ] }, { "cell_type": "code", "execution_count": 6, "id": "d7efae06-415c-425d-a0b2-bb5a86ed2b16", "metadata": {}, "outputs": [], "source": [ "# observations (always the same), simulated decisions, sample values for temperature and volatility\n", "responses = []\n", "for i in range(N):\n", " # set the tonic volatility for this agent and run the perceptual model forward\n", " agent.attributes[2][\"tonic_volatility\"] = volatilities[i]\n", " agent.input_data(input_data=u)\n", "\n", " # get decision probabilities using the belief trajectories\n", " # and the sigmoid decision function with inverse temperature\n", " p = sigmoid_inverse_temperature(\n", " x=agent.node_trajectories[1][\"expected_mean\"], temperature=temperatures[i]\n", " )\n", "\n", " # save the observations and decisions separately\n", " responses.append(np.random.binomial(p=p, n=1))" ] }, { "cell_type": "markdown", "id": "b2f1c173-ebf9-4f65-9003-a07b547bc670", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "source": [ "## Inference from the simulated behaviours" ] }, { "cell_type": "code", "execution_count": 7, "id": "92a9d137-7c0c-4dd7-bdd0-403dde0a6f90", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "outputs": [], "source": [ "hgf_logp_op = HGFDistribution(\n", " n_levels=2,\n", " model_type=\"binary\",\n", " input_data=jnp.array(\n", " [u] * N\n", " ), # the inputs are the same for all agents - just duplicate the array\n", " response_function=binary_softmax_inverse_temperature,\n", " response_function_inputs=jnp.array(responses),\n", ")" ] }, { "cell_type": "markdown", "id": "fd8b6cc5-2eca-4d5f-89e4-eb450e0832e8", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "source": [ "Here, we are not assuming hyperpriors to ensure that individual estimates are independent and avoid hierarchical partial pooling." ] }, { "cell_type": "code", "execution_count": 8, "id": "2e4177b3-2e49-43f8-9f3b-64226420ef05", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "outputs": [], "source": [ "with pm.Model() as two_levels_binary_hgf:\n", "\n", " # tonic volatility\n", " volatility = pm.Normal.dist(-3.0, 5, shape=N)\n", " censored_volatility = pm.Censored(\n", " \"censored_volatility\", volatility, lower=-8, upper=2\n", " )\n", "\n", " # inverse temperature\n", " inverse_temperature = pm.Uniform(\n", " \"inverse_temperature\", 0.2, 20, shape=N, initval=np.ones(N)\n", " )\n", "\n", " # The multi-HGF distribution\n", " # --------------------------\n", " pm.Potential(\n", " \"hgf_loglike\",\n", " hgf_logp_op(\n", " tonic_volatility_2=censored_volatility,\n", " response_function_parameters=inverse_temperature,\n", " ),\n", " )" ] }, { "cell_type": "code", "execution_count": 9, "id": "5154c464-7218-4ebf-a974-0fbaf82685d5", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "Auto-assigning NUTS sampler...\n", "Initializing NUTS using jitter+adapt_diag...\n", "Sequential sampling (2 chains in 1 job)\n", "NUTS: [censored_volatility, inverse_temperature]\n" ] }, { "data": { "application/vnd.jupyter.widget-view+json": { "model_id": "8539e39ed4cd43bc843ef92c97797701", "version_major": 2, "version_minor": 0 }, "text/plain": [ "Output()" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "
\n"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "
\n",
       "
\n" ], "text/plain": [ "\n" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "application/vnd.jupyter.widget-view+json": { "model_id": "053487bfb1a748e4a5f00a76d578e03a", "version_major": 2, "version_minor": 0 }, "text/plain": [ "Output()" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "
\n"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "
\n",
       "
\n" ], "text/plain": [ "\n" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stderr", "output_type": "stream", "text": [ "Sampling 2 chains for 1_000 tune and 1_000 draw iterations (2_000 + 2_000 draws total) took 70 seconds.\n", "There were 2000 divergences after tuning. Increase `target_accept` or reparameterize.\n", "We recommend running at least 4 chains for robust computation of convergence diagnostics\n", "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n", "The effective sample size per chain is smaller than 100 for some parameters. A higher number is needed for reliable rhat and ess computation. See https://arxiv.org/abs/1903.08008 for details\n" ] } ], "source": [ "with two_levels_binary_hgf:\n", " two_level_hgf_idata = pm.sample(chains=2, cores=1)" ] }, { "cell_type": "markdown", "id": "51c38758-8394-40ca-ae32-676d44f59dd4", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "source": [ "## Visualizing parameters recovery" ] }, { "cell_type": "markdown", "id": "5ba30cbb-bf8c-4e00-9198-0bbe411daeae", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "source": [ "A successful parameter recovery is usually inferred from the scatterplot of simulated values and inferred values of the parameters. Here, we can see that the model can recover fairly accurate values close to the underlying parameters. Additionally, we can report the coefficient of correlation between the two variables, as a more objective measure of correspondence." ] }, { "cell_type": "code", "execution_count": 10, "id": "e8203a17-e1a2-4d32-a477-d2bd55e22ba6", "metadata": { "editable": true, "scrolled": true, "slideshow": { "slide_type": "" }, "tags": [ "hide-input" ] }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "fig, axs = plt.subplots(figsize=(12, 5), ncols=2)\n", "\n", "axs[0].plot([-6.0, 0.0], [-6.0, 0.0], color=\"grey\", linestyle=\"--\", zorder=-1)\n", "axs[1].plot([0.0, 7.0], [0.0, 7.0], color=\"grey\", linestyle=\"--\", zorder=-1)\n", "\n", "for var_name, refs, idx in zip(\n", " [\"censored_volatility\", \"inverse_temperature\"],\n", " [volatilities, temperatures],\n", " [0, 1],\n", "):\n", " inferred_parameters = az.summary(two_level_hgf_idata, var_names=var_name)[\n", " \"mean\"\n", " ].tolist()\n", "\n", " sns.kdeplot(\n", " x=refs,\n", " y=inferred_parameters,\n", " ax=axs[idx],\n", " fill=True,\n", " cmap=\"Reds\" if var_name == \"censored_volatility\" else \"Greens\",\n", " alpha=0.4,\n", " )\n", "\n", " axs[idx].scatter(\n", " refs,\n", " az.summary(two_level_hgf_idata, var_names=var_name)[\"mean\"].tolist(),\n", " s=60,\n", " alpha=0.8,\n", " edgecolors=\"k\",\n", " color=\"#c44e52\" if var_name == \"censored_volatility\" else \"#55a868\",\n", " )\n", "\n", " axs[idx].grid(True, linestyle=\"--\")\n", " axs[idx].set_xlabel(\"Simulated parameter\")\n", " axs[idx].set_ylabel(\"Infered parameter\")\n", "\n", "\n", "axs[0].set_title(\"Second level tonic volatility\")\n", "axs[1].set_title(\"Inverse temperature\")\n", "sns.despine()" ] }, { "cell_type": "markdown", "id": "e4cf5f0c-6263-45a1-860a-6694b9a02279", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "source": [ "# System configuration" ] }, { "cell_type": "code", "execution_count": 11, "id": "f26d0d76-8705-4f03-86fc-e8d1b996b99c", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Last updated: Thu Jun 27 2024\n", "\n", "Python implementation: CPython\n", "Python version : 3.12.3\n", "IPython version : 8.23.0\n", "\n", "pyhgf : 0.1.2\n", "jax : 0.4.30\n", "jaxlib: 0.4.30\n", "\n", "numpy : 1.26.0\n", "matplotlib: 3.8.4\n", "sys : 3.12.3 | packaged by conda-forge | (main, Apr 15 2024, 18:38:13) [GCC 12.3.0]\n", "jax : 0.4.30\n", "pymc : 5.16.1\n", "arviz : 0.18.0\n", "seaborn : 0.13.2\n", "\n", "Watermark: 2.4.3\n", "\n" ] } ], "source": [ "%load_ext watermark\n", "%watermark -n -u -v -iv -w -p pyhgf,jax,jaxlib" ] }, { "cell_type": "code", "execution_count": null, "id": "1467d1a4-9bc8-4cf9-9b71-85c797d7973e", "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "interpreter": { "hash": "40d3a090f54c6569ab1632332b64b2c03c39dcf918b08424e98f38b5ae0af88f" }, "kernelspec": { "display_name": "Python 3 (ipykernel)", "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": "3.12.3" } }, "nbformat": 4, "nbformat_minor": 5 }