The Tensorflow Inception model has been trained to recognize objects of 1000 classes. If you feed an image to the network it will spit out the likelihood of each class for the object shown in the image. You can get these files by downloading and unzipping'inception5h.zip' First of all we have to know, that the Tensorflow Inception net accepts 224x224 sized input images. That's the reason why we resize the image such that it's largest dimension is 224 and we pad the image's remaining dimension with white pixels, such that the width height (padToSquare).

Because chances are it can learn to outsmart you inside a day. Earlier this year, we reported that Alphabet's machine-learning subsidiary, DeepMind, had made a huge advance. Using an artificial-intelligence approach known as reinforcement learning, it had enabled its AlphaGo software to develop superhuman skills for the game of Go without needing human data. Armed with just the rules of the game, the AI was able to make random plays until it developed champion-beating strategies. The new software was dubbed AlphaGo Zero because it didn't need any human input. Now, in a paper published on arXiv, the DeepMind team reports that the software has been generalized so that it can learn other games.

My colleague Sandy Carter delivered the Enterprise Innovation State of the Union last week at AWS re:Invent. She wrote the guest post below to recap the announcements that she made from the stage. "I want my company to innovate, but I am not convinced we can execute successfully." Far too many times I have heard this fear expressed by senior executives that I have met at different points in my career. In fact, a recent study published by Price Waterhouse Coopers found that while 93% of executives depend on innovation to drive growth, more than half are challenged to take innovative ideas to market quickly in a scalable way.

Progress in deep learning is slowed by the days or weeks it takes to train large models. The natural solution of using more hardware is limited by diminishing returns, and leads to inefficient use of additional resources. In this paper, we present a large batch, stochastic optimization algorithm that is both faster than widely used algorithms for fixed amounts of computation, and also scales up substantially better as more computational resources become available. Our algorithm implicitly computes the inverse Hessian of each mini-batch to produce descent directions; we do so without either an explicit approximation to the Hessian or Hessian-vector products. We demonstrate the effectiveness of our algorithm by successfully training large ImageNet models (Inception-V3, Resnet-50, Resnet-101 and Inception-Resnet-V2) with mini-batch sizes of up to 32000 with no loss in validation error relative to current baselines, and no increase in the total number of steps. At smaller mini-batch sizes, our optimizer improves the validation error in these models by 0.8-0.9%. Alternatively, we can trade off this accuracy to reduce the number of training steps needed by roughly 10-30%. Our work is practical and easily usable by others -- only one hyperparameter (learning rate) needs tuning, and furthermore, the algorithm is as computationally cheap as the commonly used Adam optimizer.

Direct acoustics-to-word (A2W) models in the end-to-end paradigm have received increasing attention compared to conventional sub-word based automatic speech recognition models using phones, characters, or context-dependent hidden Markov model states. This is because A2W models recognize words from speech without any decoder, pronunciation lexicon, or externally-trained language model, making training and decoding with such models simple. Prior work has shown that A2W models require orders of magnitude more training data in order to perform comparably to conventional models. Our work also showed this accuracy gap when using the English Switchboard-Fisher data set. This paper describes a recipe to train an A2W model that closes this gap and is at-par with state-of-the-art sub-word based models. We achieve a word error rate of 8.8%/13.9% on the Hub5-2000 Switchboard/CallHome test sets without any decoder or language model. We find that model initialization, training data order, and regularization have the most impact on the A2W model performance. Next, we present a joint word-character A2W model that learns to first spell the word and then recognize it. This model provides a rich output to the user instead of simple word hypotheses, making it especially useful in the case of words unseen or rarely-seen during training.

Current deep learning results on video generation are limited while there are only a few first results on video prediction and no relevant significant results on video completion. This is due to the severe ill-posedness inherent in these three problems. In this paper, we focus on human action videos, and propose a general, two-stage deep framework to generate human action videos with no constraints or arbitrary number of constraints, which uniformly address the three problems: video generation given no input frames, video prediction given the first few frames, and video completion given the first and last frames. To make the problem tractable, in the first stage we train a deep generative model that generates a human pose sequence from random noise. In the second stage, a skeleton-to-image network is trained, which is used to generate a human action video given the complete human pose sequence generated in the first stage. By introducing the two-stage strategy, we sidestep the original ill-posed problems while producing for the first time high-quality video generation/prediction/completion results of much longer duration. We present quantitative and qualitative evaluation to show that our two-stage approach outperforms state-of-the-art methods in video generation, prediction and video completion. Our video result demonstration can be viewed at https://iamacewhite.github.io/supp/index.html

Although the word-popularity based negative sampler has shown superb performance in the skip-gram model, the theoretical motivation behind oversampling popular (non-observed) words as negative samples is still not well understood. In this paper, we start from an investigation of the gradient vanishing issue in the skip-gram model without a proper negative sampler. By performing an insightful analysis from the stochastic gradient descent (SGD) learning perspective, we demonstrate that, both theoretically and intuitively, negative samples with larger inner product scores are more informative than those with lower scores for the SGD learner in terms of both convergence rate and accuracy. Understanding this, we propose an alternative sampling algorithm that dynamically selects informative negative samples during each SGD update. More importantly, the proposed sampler accounts for multi-dimensional self-embedded features during the sampling process, which essentially makes it more effective than the original popularity-based (one-dimensional) sampler. Empirical experiments further verify our observations, and show that our fine-grained samplers gain significant improvement over the existing ones without increasing computational complexity.

This paper introduces a probabilistic framework for k-shot image classification. The goal is to generalise from an initial large-scale classification task to a separate task comprising new classes and small numbers of examples. The new approach not only leverages the feature-based representation learned by a neural network from the initial task (representational transfer), but also information about the classes (concept transfer). The concept information is encapsulated in a probabilistic model for the final layer weights of the neural network which acts as a prior for probabilistic k-shot learning. We show that even a simple probabilistic model achieves state-of-the-art on a standard k-shot learning dataset by a large margin. Moreover, it is able to accurately model uncertainty, leading to well calibrated classifiers, and is easily extensible and flexible, unlike many recent approaches to k-shot learning.

Summary: As a profession we do a pretty poor job of agreeing on good naming conventions for really important parts of our professional lives. "Machine Learning" is just the most recent case in point. It's had a perfectly good definition for a very long time, but now the deep learning folks are trying to hijack the term. Let's make up our minds.

Matrix multiplication is a key computation within many scientific applications, particularly those in deep learning. Many operations in modern deep neural networks are either defined as matrix multiplications or can be cast as such. As an example, the NVIDIA cuDNN library implements convolutions for neural networks using various flavors of matrix multiplication, such as the classical formulation of direct convolution as a matrix product between image-to-column and filter datasets [1]. Matrix multiplication is also the core routine when computing convolutions based on Fast Fourier Transforms (FFT) [2] or the Winograd approach [3]. When constructing cuDNN, we began with our high-performance implementations of general matrix multiplication (GEMM) in the cuBLAS library, supplementing and tailoring them to efficiently compute convolution. Today, our ability to adapt these GEMM strategies and algorithms is critical to delivering the best performance for many different problems and applications within deep learning.