Paper Group AWR 116
Towards Interactive Training of Non-Player Characters in Video Games. Counterpoint by Convolution. Incremental Learning Techniques for Semantic Segmentation. Graph Interpolating Activation Improves Both Natural and Robust Accuracies in Data-Efficient Deep Learning. How to Initialize your Network? Robust Initialization for WeightNorm & ResNets. Anti …
Towards Interactive Training of Non-Player Characters in Video Games
| Title | Towards Interactive Training of Non-Player Characters in Video Games |
| Authors | Igor Borovikov, Jesse Harder, Michael Sadovsky, Ahmad Beirami |
| Abstract | There is a high demand for high-quality Non-Player Characters (NPCs) in video games. Hand-crafting their behavior is a labor intensive and error prone engineering process with limited controls exposed to the game designers. We propose to create such NPC behaviors interactively by training an agent in the target environment using imitation learning with a human in the loop. While traditional behavior cloning may fall short of achieving the desired performance, we show that interactivity can substantially improve it with a modest amount of human efforts. The model we train is a multi-resolution ensemble of Markov models, which can be used as is or can be further “compressed” into a more compact model for inference on consumer devices. We illustrate our approach on an example in OpenAI Gym, where a human can help to quickly train an agent with only a handful of interactive demonstrations. We also outline our experiments with NPC training for a first-person shooter game currently in development. |
| Tasks | Imitation Learning |
| Published | 2019-06-03 |
| URL | https://arxiv.org/abs/1906.00535v1 |
| https://arxiv.org/pdf/1906.00535v1.pdf | |
| PWC | https://paperswithcode.com/paper/190600535 |
| Repo | https://github.com/nekkar/interactive_training |
| Framework | none |
Counterpoint by Convolution
| Title | Counterpoint by Convolution |
| Authors | Cheng-Zhi Anna Huang, Tim Cooijmans, Adam Roberts, Aaron Courville, Douglas Eck |
| Abstract | Machine learning models of music typically break up the task of composition into a chronological process, composing a piece of music in a single pass from beginning to end. On the contrary, human composers write music in a nonlinear fashion, scribbling motifs here and there, often revisiting choices previously made. In order to better approximate this process, we train a convolutional neural network to complete partial musical scores, and explore the use of blocked Gibbs sampling as an analogue to rewriting. Neither the model nor the generative procedure are tied to a particular causal direction of composition. Our model is an instance of orderless NADE (Uria et al., 2014), which allows more direct ancestral sampling. However, we find that Gibbs sampling greatly improves sample quality, which we demonstrate to be due to some conditional distributions being poorly modeled. Moreover, we show that even the cheap approximate blocked Gibbs procedure from Yao et al. (2014) yields better samples than ancestral sampling, based on both log-likelihood and human evaluation. |
| Tasks | |
| Published | 2019-03-18 |
| URL | http://arxiv.org/abs/1903.07227v1 |
| http://arxiv.org/pdf/1903.07227v1.pdf | |
| PWC | https://paperswithcode.com/paper/counterpoint-by-convolution |
| Repo | https://github.com/kevindonoghue/coconet-pytorch |
| Framework | pytorch |
Incremental Learning Techniques for Semantic Segmentation
| Title | Incremental Learning Techniques for Semantic Segmentation |
| Authors | Umberto Michieli, Pietro Zanuttigh |
| Abstract | Deep learning architectures exhibit a critical drop of performance due to catastrophic forgetting when they are required to incrementally learn new tasks. Contemporary incremental learning frameworks focus on image classification and object detection while in this work we formally introduce the incremental learning problem for semantic segmentation in which a pixel-wise labeling is considered. To tackle this task we propose to distill the knowledge of the previous model to retain the information about previously learned classes, whilst updating the current model to learn the new ones. We propose various approaches working both on the output logits and on intermediate features. In opposition to some recent frameworks, we do not store any image from previously learned classes and only the last model is needed to preserve high accuracy on these classes. The experimental evaluation on the Pascal VOC2012 dataset shows the effectiveness of the proposed approaches. |
| Tasks | Image Classification, Object Detection, Semantic Segmentation |
| Published | 2019-07-31 |
| URL | https://arxiv.org/abs/1907.13372v4 |
| https://arxiv.org/pdf/1907.13372v4.pdf | |
| PWC | https://paperswithcode.com/paper/incremental-learning-techniques-for-semantic |
| Repo | https://github.com/LTTM/IL-SemSegm |
| Framework | tf |
Graph Interpolating Activation Improves Both Natural and Robust Accuracies in Data-Efficient Deep Learning
| Title | Graph Interpolating Activation Improves Both Natural and Robust Accuracies in Data-Efficient Deep Learning |
| Authors | Bao Wang, Stanley J. Osher |
| Abstract | Improving the accuracy and robustness of deep neural nets (DNNs) and adapting them to small training data are primary tasks in deep learning research. In this paper, we replace the output activation function of DNNs, typically the data-agnostic softmax function, with a graph Laplacian-based high dimensional interpolating function which, in the continuum limit, converges to the solution of a Laplace-Beltrami equation on a high dimensional manifold. Furthermore, we propose end-to-end training and testing algorithms for this new architecture. The proposed DNN with graph interpolating activation integrates the advantages of both deep learning and manifold learning. Compared to the conventional DNNs with the softmax function as output activation, the new framework demonstrates the following major advantages: First, it is better applicable to data-efficient learning in which we train high capacity DNNs without using a large number of training data. Second, it remarkably improves both natural accuracy on the clean images and robust accuracy on the adversarial images crafted by both white-box and black-box adversarial attacks. Third, it is a natural choice for semi-supervised learning. For reproducibility, the code is available at \url{https://github.com/BaoWangMath/DNN-DataDependentActivation}. |
| Tasks | |
| Published | 2019-07-16 |
| URL | https://arxiv.org/abs/1907.06800v1 |
| https://arxiv.org/pdf/1907.06800v1.pdf | |
| PWC | https://paperswithcode.com/paper/graph-interpolating-activation-improves-both |
| Repo | https://github.com/BaoWangMath/DNN-DataDependentActivation |
| Framework | pytorch |
How to Initialize your Network? Robust Initialization for WeightNorm & ResNets
| Title | How to Initialize your Network? Robust Initialization for WeightNorm & ResNets |
| Authors | Devansh Arpit, Victor Campos, Yoshua Bengio |
| Abstract | Residual networks (ResNet) and weight normalization play an important role in various deep learning applications. However, parameter initialization strategies have not been studied previously for weight normalized networks and, in practice, initialization methods designed for un-normalized networks are used as a proxy. Similarly, initialization for ResNets have also been studied for un-normalized networks and often under simplified settings ignoring the shortcut connection. To address these issues, we propose a novel parameter initialization strategy that avoids explosion/vanishment of information across layers for weight normalized networks with and without residual connections. The proposed strategy is based on a theoretical analysis using mean field approximation. We run over 2,500 experiments and evaluate our proposal on image datasets showing that the proposed initialization outperforms existing initialization methods in terms of generalization performance, robustness to hyper-parameter values and variance between seeds, especially when networks get deeper in which case existing methods fail to even start training. Finally, we show that using our initialization in conjunction with learning rate warmup is able to reduce the gap between the performance of weight normalized and batch normalized networks. |
| Tasks | |
| Published | 2019-06-05 |
| URL | https://arxiv.org/abs/1906.02341v2 |
| https://arxiv.org/pdf/1906.02341v2.pdf | |
| PWC | https://paperswithcode.com/paper/how-to-initialize-your-network-robust |
| Repo | https://github.com/victorcampos7/weightnorm-init |
| Framework | pytorch |
Antipodal Robotic Grasping using Generative Residual Convolutional Neural Network
| Title | Antipodal Robotic Grasping using Generative Residual Convolutional Neural Network |
| Authors | Sulabh Kumra, Shirin Joshi, Ferat Sahin |
| Abstract | In this paper, we present a modular robotic system to tackle the problem of generating and performing antipodal robotic grasps for unknown objects from n-channel image of the scene. We propose a novel Generative Residual Convolutional Neural Network (GR-ConvNet) model that can generate robust antipodal grasps from n-channel input at real-time speeds (~20ms). We evaluate the proposed model architecture on standard datasets and a diverse set of household objects. We achieved state-of-the-art accuracy of 97.7% and 94.6% on Cornell and Jacquard grasping datasets respectively. We also demonstrate a grasp success rate of 95.4% and 93% on household and adversarial objects respectively using a 7 DoF robotic arm. |
| Tasks | Robotic Grasping |
| Published | 2019-09-11 |
| URL | https://arxiv.org/abs/1909.04810v2 |
| https://arxiv.org/pdf/1909.04810v2.pdf | |
| PWC | https://paperswithcode.com/paper/antipodal-robotic-grasping-using-generative |
| Repo | https://github.com/skumra/robotic-grasping |
| Framework | pytorch |
Convex Covariate Clustering for Classification
| Title | Convex Covariate Clustering for Classification |
| Authors | Daniel Andrade, Kenji Fukumizu, Yuzuru Okajima |
| Abstract | Clustering, like covariate selection for classification, is an important step to understand and interpret the data. However, clustering of covariates is often performed independently of the classification step, which can lead to undesirable clustering results. Therefore, we propose a method that can cluster covariates while taking into account class label information of samples. We formulate the problem as a convex optimization problem which uses both, a-priori similarity information between covariates, and information from class-labeled samples. Like convex clustering [Chi and Lange, 2015], the proposed method offers a unique global minima making it insensitive to initialization. In order to solve the convex problem, we propose a specialized alternating direction method of multipliers (ADMM), which scales up to several thousands of variables. Furthermore, in order to circumvent computationally expensive cross-validation, we propose a model selection criterion based on approximate marginal likelihood estimation. Experiments on synthetic and real data confirm the usefulness of the proposed clustering method and the selection criterion. |
| Tasks | Model Selection |
| Published | 2019-03-05 |
| URL | http://arxiv.org/abs/1903.01680v1 |
| http://arxiv.org/pdf/1903.01680v1.pdf | |
| PWC | https://paperswithcode.com/paper/convex-covariate-clustering-for |
| Repo | https://github.com/andrade-stats/convexCovariateClusteringClassification |
| Framework | none |
Macaw: An Extensible Conversational Information Seeking Platform
| Title | Macaw: An Extensible Conversational Information Seeking Platform |
| Authors | Hamed Zamani, Nick Craswell |
| Abstract | Conversational information seeking (CIS) has been recognized as a major emerging research area in information retrieval. Such research will require data and tools, to allow the implementation and study of conversational systems. This paper introduces Macaw, an open-source framework with a modular architecture for CIS research. Macaw supports multi-turn, multi-modal, and mixed-initiative interactions, and enables research for tasks such as document retrieval, question answering, recommendation, and structured data exploration. It has a modular design to encourage the study of new CIS algorithms, which can be evaluated in batch mode. It can also integrate with a user interface, which allows user studies and data collection in an interactive mode, where the back end can be fully algorithmic or a wizard of oz setup. Macaw is distributed under the MIT License. |
| Tasks | Information Retrieval, Question Answering |
| Published | 2019-12-18 |
| URL | https://arxiv.org/abs/1912.08904v1 |
| https://arxiv.org/pdf/1912.08904v1.pdf | |
| PWC | https://paperswithcode.com/paper/macaw-an-extensible-conversational |
| Repo | https://github.com/microsoft/macaw |
| Framework | pytorch |
Bayesian Online Detection and Prediction of Change Points
| Title | Bayesian Online Detection and Prediction of Change Points |
| Authors | Diego Agudelo-España, Sebastian Gomez-Gonzalez, Stefan Bauer, Bernhard Schölkopf, Jan Peters |
| Abstract | Online detection of instantaneous changes in the generative process of a data sequence generally focuses on retrospective inference of such change points without considering their future occurrences. We extend the Bayesian Online Change Point Detection algorithm to also infer the number of time steps until the next change point (i.e., the residual time). This enables us to handle observation models which depend on the total segment duration, which is useful to model data sequences with temporal scaling. In addition, we extend the model by removing the i.i.d. assumption on the observation model parameters. The resulting inference algorithm for segment detection can be deployed in an online fashion, and we illustrate applications to synthetic and to two medical real-world data sets. |
| Tasks | Bayesian Inference, Change Point Detection |
| Published | 2019-02-12 |
| URL | http://arxiv.org/abs/1902.04524v1 |
| http://arxiv.org/pdf/1902.04524v1.pdf | |
| PWC | https://paperswithcode.com/paper/bayesian-online-detection-and-prediction-of |
| Repo | https://github.com/DiegoAE/BOSD |
| Framework | none |
Attention-Augmented End-to-End Multi-Task Learning for Emotion Prediction from Speech
| Title | Attention-Augmented End-to-End Multi-Task Learning for Emotion Prediction from Speech |
| Authors | Zixing Zhang, Bingwen Wu, Bjoern Schuller |
| Abstract | Despite the increasing research interest in end-to-end learning systems for speech emotion recognition, conventional systems either suffer from the overfitting due in part to the limited training data, or do not explicitly consider the different contributions of automatically learnt representations for a specific task. In this contribution, we propose a novel end-to-end framework which is enhanced by learning other auxiliary tasks and an attention mechanism. That is, we jointly train an end-to-end network with several different but related emotion prediction tasks, i.e., arousal, valence, and dominance predictions, to extract more robust representations shared among various tasks than traditional systems with the hope that it is able to relieve the overfitting problem. Meanwhile, an attention layer is implemented on top of the layers for each task, with the aim to capture the contribution distribution of different segment parts for each individual task. To evaluate the effectiveness of the proposed system, we conducted a set of experiments on the widely used database IEMOCAP. The empirical results show that the proposed systems significantly outperform corresponding baseline systems. |
| Tasks | Emotion Recognition, Multi-Task Learning, Speech Emotion Recognition |
| Published | 2019-03-29 |
| URL | http://arxiv.org/abs/1903.12424v1 |
| http://arxiv.org/pdf/1903.12424v1.pdf | |
| PWC | https://paperswithcode.com/paper/attention-augmented-end-to-end-multi-task |
| Repo | https://github.com/16631140828/Paper-list |
| Framework | none |
What’s Hidden in a Randomly Weighted Neural Network?
| Title | What’s Hidden in a Randomly Weighted Neural Network? |
| Authors | Vivek Ramanujan, Mitchell Wortsman, Aniruddha Kembhavi, Ali Farhadi, Mohammad Rastegari |
| Abstract | Training a neural network is synonymous with learning the values of the weights. By contrast, we demonstrate that randomly weighted neural networks contain subnetworks which achieve impressive performance without ever training the weight values. Hidden in a randomly weighted Wide ResNet-50 we show that there is a subnetwork (with random weights) that is smaller than, but matches the performance of a ResNet-34 trained on ImageNet. Not only do these “untrained subnetworks” exist, but we provide an algorithm to effectively find them. We empirically show that as randomly weighted neural networks with fixed weights grow wider and deeper, an “untrained subnetwork” approaches a network with learned weights in accuracy. Our code and pretrained models are available at https://github.com/allenai/hidden-networks. |
| Tasks | Image Classification |
| Published | 2019-11-29 |
| URL | https://arxiv.org/abs/1911.13299v2 |
| https://arxiv.org/pdf/1911.13299v2.pdf | |
| PWC | https://paperswithcode.com/paper/whats-hidden-in-a-randomly-weighted-neural |
| Repo | https://github.com/allenai/hidden-networks |
| Framework | pytorch |
A BERT Baseline for the Natural Questions
| Title | A BERT Baseline for the Natural Questions |
| Authors | Chris Alberti, Kenton Lee, Michael Collins |
| Abstract | This technical note describes a new baseline for the Natural Questions. Our model is based on BERT and reduces the gap between the model F1 scores reported in the original dataset paper and the human upper bound by 30% and 50% relative for the long and short answer tasks respectively. This baseline has been submitted to the official NQ leaderboard at ai.google.com/research/NaturalQuestions. Code, preprocessed data and pretrained model are available at https://github.com/google-research/language/tree/master/language/question_answering/bert_joint. |
| Tasks | Question Answering |
| Published | 2019-01-24 |
| URL | https://arxiv.org/abs/1901.08634v3 |
| https://arxiv.org/pdf/1901.08634v3.pdf | |
| PWC | https://paperswithcode.com/paper/a-bert-baseline-for-the-natural-questions |
| Repo | https://github.com/gooofy/zbrain |
| Framework | tf |
Structural Language Models of Code
| Title | Structural Language Models of Code |
| Authors | Uri Alon, Roy Sadaka, Omer Levy, Eran Yahav |
| Abstract | We address the problem of any-code completion - generating a missing piece of source code in a given program without any restriction on the vocabulary or structure. We introduce a new approach to any-code completion that leverages the strict syntax of programming languages to model a code snippet as a tree - structural language modeling (SLM). SLM estimates the probability of the program’s abstract syntax tree (AST) by decomposing it into a product of conditional probabilities over its nodes. We present a neural model that computes these conditional probabilities by considering all AST paths leading to a target node. Unlike previous techniques that have severely restricted the kinds of expressions that can be generated in this task, our approach can generate arbitrary code in any programming language. Our model significantly outperforms both seq2seq and a variety of structured approaches in generating Java and C# code. We make our code, datasets, and models publicly available. |
| Tasks | Code Generation, Language Modelling |
| Published | 2019-09-30 |
| URL | https://arxiv.org/abs/1910.00577v2 |
| https://arxiv.org/pdf/1910.00577v2.pdf | |
| PWC | https://paperswithcode.com/paper/structural-language-models-for-any-code |
| Repo | https://github.com/tech-srl/code2vec |
| Framework | tf |
Improved memory in recurrent neural networks with sequential non-normal dynamics
| Title | Improved memory in recurrent neural networks with sequential non-normal dynamics |
| Authors | A. Emin Orhan, Xaq Pitkow |
| Abstract | Training recurrent neural networks (RNNs) is a hard problem due to degeneracies in the optimization landscape, a problem also known as vanishing/exploding gradients. Short of designing new RNN architectures, previous methods for dealing with this problem usually boil down to orthogonalization of the recurrent dynamics, either at initialization or during the entire training period. The basic motivation behind these methods is that orthogonal transformations are isometries of the Euclidean space, hence they preserve (Euclidean) norms and effectively deal with vanishing/exploding gradients. However, this ignores the crucial effects of non-linearity and noise. In the presence of a non-linearity, orthogonal transformations no longer preserve norms, suggesting that alternative transformations might be better suited to non-linear networks. Moreover, in the presence of noise, norm preservation itself ceases to be the ideal objective. A more sensible objective is maximizing the signal-to-noise ratio (SNR) of the propagated signal instead. Previous work has shown that in the linear case, recurrent networks that maximize the SNR display strongly non-normal, sequential dynamics and orthogonal networks are highly suboptimal by this measure. Motivated by this finding, here we investigate the potential of non-normal RNNs, i.e. RNNs with a non-normal recurrent connectivity matrix, in sequential processing tasks. Our experimental results show that non-normal RNNs outperform their orthogonal counterparts in a diverse range of benchmarks. We also find evidence for increased non-normality and hidden chain-like feedforward motifs in trained RNNs initialized with orthogonal recurrent connectivity matrices. |
| Tasks | |
| Published | 2019-05-31 |
| URL | https://arxiv.org/abs/1905.13715v2 |
| https://arxiv.org/pdf/1905.13715v2.pdf | |
| PWC | https://paperswithcode.com/paper/improved-memory-in-recurrent-neural-networks |
| Repo | https://github.com/eminorhan/nonnormal-init |
| Framework | pytorch |
Dynamic Mode Decomposition based feature for Image Classification
| Title | Dynamic Mode Decomposition based feature for Image Classification |
| Authors | Rahul-Vigneswaran K, Sachin-Kumar S, Neethu Mohan, Soman KP |
| Abstract | Irrespective of the fact that Machine learning has produced groundbreaking results, it demands an enormous amount of data in order to perform so. Even though data production has been in its all-time high, almost all the data is unlabelled, hence making them unsuitable for training the algorithms. This paper proposes a novel method of extracting the features using Dynamic Mode Decomposition (DMD). The experiment is performed using data samples from Imagenet. The learning is done using SVM-linear, SVM-RBF, Random Kitchen Sink approach (RKS). The results have shown that DMD features with RKS give competing results. |
| Tasks | Image Classification |
| Published | 2019-10-08 |
| URL | https://arxiv.org/abs/1910.03188v1 |
| https://arxiv.org/pdf/1910.03188v1.pdf | |
| PWC | https://paperswithcode.com/paper/dynamic-mode-decomposition-based-feature-for |
| Repo | https://github.com/rahulvigneswaran/Dynamic-Mode-Decomposition-based-feature-for-Image-Classification |
| Framework | none |
