The effort devoted to hand-crafting neural network image classifiers has motivated the use of architecture search to discover them automatically. Although evolutionary algorithms have been repeatedly applied to neural network topologies, the image classifiers thus discovered have remained inferior to human-crafted ones. Here, we evolve an image classifier-AmoebaNet-A-that surpasses hand-designs for the first time.To do this, we modify the tournament selection evolutionary algorithm by introducing an age property to favor the younger genotypes. Matching size, AmoebaNet-A has comparable accuracy to current state-of-the-art ImageNet models discovered with more complex architecture-search methods. Scaled to larger size, AmoebaNet-A sets a new state-of-theart 83.9% top-1 / 96.6% top-5 ImageNet accuracy. In a controlled comparison against a well known reinforcement learning algorithm, we give evidence that evolution can obtain results faster with the same hardware, especially at the earlier stages of the search. This is relevant when fewer compute resources are available. Evolution is, thus, a simple method to effectively discover high-quality architectures.Review papers provide informative surveys of earlier [18,49] and more recent [15] literature on image classifier architecture search, including successful RL studies [2,6,29,[52][53][54] and evolutionary studies like those mentioned in 1 After our submission, a recent preprint has further scaled up and retrained AmoebaNet-A to reach 84.3% top-1 / 97.0% top-5 ImageNet accuracy [25].

Predictive coding is a unifying framework for understanding redundancy reduction and efficient coding in the nervous system. By transmitting only the unpredicted portions of an incoming sensory signal, predictive coding allows the nervous system to reduce redundancy and make full use of the limited dynamic range of neurons. Starting with the hypothesis of efficient coding as a design principle in the sensory system, predictive coding provides a functional explanation for a range of neural responses and many aspects of brain organization. The lateral and temporal antagonism in receptive fields in the retina and lateral geniculate nucleus occur naturally as a consequence of predictive coding of natural images. In the higher visual system, predictive coding provides an explanation for oriented receptive fields and contextual effects as well as the hierarchical reciprocally connected organization of the cortex. Predictive coding has also been found to be consistent with a variety of neurophysiological and psychophysical data obtained from different areas of the brain. WIREs Cogni Sci 2011 2 580-593 DOI: 10.1002/wcs.142 For further resources related to this article, please visit the WIREs website.

Neural network scaling has been critical for improving the model quality in many real-world machine learning applications with vast amounts of training data and compute. Although this trend of scaling is affirmed to be a sure-fire approach for better model quality, there are challenges on the path such as the computation cost, ease of programming, and efficient implementation on parallel devices. GShard is a module composed of a set of lightweight annotation APIs and an extension to the XLA compiler. It provides an elegant way to express a wide range of parallel computation patterns with minimal changes to the existing model code. GShard enabled us to scale up multilingual neural machine translation Transformer model with Sparsely-Gated Mixture-of-Experts beyond 600 billion parameters using automatic sharding. We demonstrate that such a giant model can efficienctly be trained on 2048 TPU v3 accelerators in 4 days to achieve far superior quality for translation from 100 languages to English compared to the prior art.Preprint. Under review.

No abstract

A key problem in neuroscience is understanding how the brain makes decisions under uncertainty. Important insights have been gained using tasks such as the random dots motion discrimination task in which the subject makes decisions based on noisy stimuli. A descriptive model known as the drift diffusion model has previously been used to explain psychometric and reaction time data from such tasks but to fully explain the data, one is forced to make ad-hoc assumptions such as a time-dependent collapsing decision boundary. We show that such assumptions are unnecessary when decision making is viewed within the framework of partially observable Markov decision processes (POMDPs). We propose an alternative model for decision making based on POMDPs. We show that the motion discrimination task reduces to the problems of (1) computing beliefs (posterior distributions) over the unknown direction and motion strength from noisy observations in a Bayesian manner, and (2) selecting actions based on these beliefs to maximize the expected sum of future rewards. The resulting optimal policy (belief-to-action mapping) is shown to be equivalent to a collapsing decision threshold that governs the switch from evidence accumulation to a discrimination decision. We show that the model accounts for both accuracy and reaction time as a function of stimulus strength as well as different speed-accuracy conditions in the random dots task.