اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدReinforcement and inference in cross-situational word learning
68
0
0.0
(
0
)
اسأل ChatGPT حول البحث
ﻻ يوجد ملخص باللغة العربية
Cross-situational word learning is based on the notion that a learner can determine the referent of a word by finding something in common across many observed uses of that word. Here we propose an adaptive learning algorithm that contains a parameter that controls the strength of the reinforcement applied to associations between concurrent words and referents, and a parameter that regulates inference, which includes built-in biases, such as mutual exclusivity, and information of past learning events. By adjusting these parameters so that the model predictions agree with data from representative experiments on cross-situational word learning, we were able to explain the learning strategies adopted by the participants of those experiments in terms of a trade-off between reinforcement and inference. These strategies can vary wildly depending on the conditions of the experiments. For instance, for fast mapping experiments (i.e., the correct referent could, in principle, be inferred in a single observation) inference is prevalent, whereas for segregated contextual diversity experiments (i.e., the referents are separated in groups and are exhibited with members of their groups only) reinforcement is predominant. Other experiments are explained with more balanced doses of reinforcement and inference.
قيم البحث
اقرأ أيضاً
An explanation for the acquisition of word-object mappings is the associative learning in a cross-situational scenario. Here we present analytical results of the performance of a simple associative learning algorithm for acquiring a one-to-one mappin
g between $N$ objects and $N$ words based solely on the co-occurrence between objects and words. In particular, a learning trial in our learning scenario consists of the presentation of $C + 1 < N$ objects together with a target word, which refers to one of the objects in the context. We find that the learning times are distributed exponentially and the learning rates are given by $ln{[frac{N(N-1)}{C + (N-1)^{2}}]}$ in the case the $N$ target words are sampled randomly and by $frac{1}{N} ln [frac{N-1}{C}] $ in the case they follow a deterministic presentation sequence. This learning performance is much superior to those exhibited by humans and more realistic learning algorithms in cross-situational experiments. We show that introduction of discrimination limitations using Webers law and forgetting reduce the performance of the associative algorithm to the human level.
The Bayesian view of the brain hypothesizes that the brain constructs a generative model of the world, and uses it to make inferences via Bayes rule. Although many types of approximate inference schemes have been proposed for hierarchical Bayesian mo
dels of the brain, the questions of how these distinct inference procedures can be realized by hierarchical networks of spiking neurons remains largely unresolved. Based on a previously proposed multi-compartment neuron model in which dendrites perform logarithmic compression, and stochastic spiking winner-take-all (WTA) circuits in which firing probability of each neuron is normalized by activities of other neurons, here we construct Spiking Neural Networks that perform emph{structured} mean-field variational inference and learning, on hierarchical directed probabilistic graphical models with discrete random variables. In these models, we do away with symmetric synaptic weights previously assumed for emph{unstructured} mean-field variational inference by learning both the feedback and feedforward weights separately. The resulting online learning rules take the form of an error-modulated local Spike-Timing-Dependent Plasticity rule. Importantly, we consider two types of WTA circuits in which only one neuron is allowed to fire at a time (hard WTA) or neurons can fire independently (soft WTA), which makes neurons in these circuits operate in regimes of temporal and rate coding respectively. We show how the hard WTA circuits can be used to perform Gibbs sampling whereas the soft WTA circuits can be used to implement a message passing algorithm that computes the marginals approximately. Notably, a simple change in the amount of lateral inhibition realizes switching between the hard and soft WTA spiking regimes. Hence the proposed network provides a unified view of the two previously disparate modes of inference and coding by spiking neurons.
Neural responses in the cortex change over time both systematically, due to ongoing plasticity and learning, and seemingly randomly, due to various sources of noise and variability. Most previous work considered each of these processes, learning and
variability, in isolation -- here we study neural networks exhibiting both and show that their interaction leads to the emergence of powerful computational properties. We trained neural networks on classical unsupervised learning tasks, in which the objective was to represent their inputs in an efficient, easily decodable form, with an additional cost for neural reliability which we derived from basic biophysical considerations. This cost on reliability introduced a tradeoff between energetically cheap but inaccurate representations and energetically costly but accurate ones. Despite the learning tasks being non-probabilistic, the networks solved this tradeoff by developing a probabilistic representation: neural variability represented samples from statistically appropriate posterior distributions that would result from performing probabilistic inference over their inputs. We provide an analytical understanding of this result by revealing a connection between the cost of reliability, and the objective for a state-of-the-art Bayesian inference strategy: variational autoencoders. We show that the same cost leads to the emergence of increasingly accurate probabilistic representations as networks become more complex, from single-layer feed-forward, through multi-layer feed-forward, to recurrent architectures. Our results provide insights into why neural responses in sensory areas show signatures of sampling-based probabilistic representations, and may inform future deep learning algorithms and their implementation in stochastic low-precision computing systems.
Active inference offers a first principle account of sentient behaviour, from which special and important cases can be derived, e.g., reinforcement learning, active learning, Bayes optimal inference, Bayes optimal design, etc. Active inference resolv
es the exploitation-exploration dilemma in relation to prior preferences, by placing information gain on the same footing as reward or value. In brief, active inference replaces value functions with functionals of (Bayesian) beliefs, in the form of an expected (variational) free energy. In this paper, we consider a sophisticated kind of active inference, using a recursive form of expected free energy. Sophistication describes the degree to which an agent has beliefs about beliefs. We consider agents with beliefs about the counterfactual consequences of action for states of affairs and beliefs about those latent states. In other words, we move from simply considering beliefs about what would happen if I did that to what would I believe about what would happen if I did that. The recursive form of the free energy functional effectively implements a deep tree search over actions and outcomes in the future. Crucially, this search is over sequences of belief states, as opposed to states per se. We illustrate the competence of this scheme, using numerical simulations of deep decision problems.
The central tenet of reinforcement learning (RL) is that agents seek to maximize the sum of cumulative rewards. In contrast, active inference, an emerging framework within cognitive and computational neuroscience, proposes that agents act to maximize
the evidence for a biased generative model. Here, we illustrate how ideas from active inference can augment traditional RL approaches by (i) furnishing an inherent balance of exploration and exploitation, and (ii) providing a more flexible conceptualization of reward. Inspired by active inference, we develop and implement a novel objective for decision making, which we term the free energy of the expected future. We demonstrate that the resulting algorithm successfully balances exploration and exploitation, simultaneously achieving robust performance on several challenging RL benchmarks with sparse, well-shaped, and no rewards.
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
