Paper Group NANR 66
FR-GAN: Fair and Robust Training. Learning Video Representations using Contrastive Bidirectional Transformer. Improved Modeling of Complex Systems Using Hybrid Physics/Machine Learning/Stochastic Models. Certified Robustness to Adversarial Label-Flipping Attacks via Randomized Smoothing. Exploration via Flow-Based Intrinsic Rewards. Robust Domain R …
FR-GAN: Fair and Robust Training
| Title | FR-GAN: Fair and Robust Training |
| Authors | Anonymous |
| Abstract | We consider the problem of fair and robust model training in the presence of data poisoning. Ensuring fairness usually involves a tradeoff against accuracy, so if the data poisoning is mistakenly viewed as additional bias to be fixed, the accuracy will be sacrificed even more. We demonstrate that this phenomenon indeed holds for state-of-the-art model fairness techniques. We then propose FR-GAN, which holistically performs fair and robust model training using generative adversarial networks (GANs). We first use a generator that attempts to classify examples as accurately as possible. In addition, we deploy two discriminators: (1) a fairness discriminator that predicts the sensitive attribute from classification results and (2) a robustness discriminator that distinguishes examples and predictions from a clean validation set. Our framework respects all the prominent fairness measures: disparate impact, equalized odds, and equal opportunity. Also, FR-GAN optimizes fairness without requiring the knowledge of prior statistics of the sensitive attributes. In our experiments, FR-GAN shows almost no decrease in fairness and accuracy in the presence of data poisoning unlike other state-of-the-art fairness methods, which are vulnerable. In addition, FR-GAN can be adjusted using parameters to maintain reasonable accuracy and fairness even if the validation set is too small or unavailable. |
| Tasks | data poisoning |
| Published | 2020-01-01 |
| URL | https://openreview.net/forum?id=BylyV1BtDB |
| https://openreview.net/pdf?id=BylyV1BtDB | |
| PWC | https://paperswithcode.com/paper/fr-gan-fair-and-robust-training |
| Repo | |
| Framework | |
Learning Video Representations using Contrastive Bidirectional Transformer
| Title | Learning Video Representations using Contrastive Bidirectional Transformer |
| Authors | Anonymous |
| Abstract | This paper proposes a self-supervised learning approach for video features that results in significantly improved performance on downstream tasks (such as video classification, captioning and segmentation) compared to existing methods. Our method extends the BERT model for text sequences to the case of sequences of real-valued feature vectors, by replacing the softmax loss with noise contrastive estimation (NCE). We also show how to learn representations from sequences of visual features and sequences of words derived from ASR (automatic speech recognition), and show that such cross-modal training (when possible) helps even more. |
| Tasks | Speech Recognition, Video Classification |
| Published | 2020-01-01 |
| URL | https://openreview.net/forum?id=rJgRMkrtDr |
| https://openreview.net/pdf?id=rJgRMkrtDr | |
| PWC | https://paperswithcode.com/paper/learning-video-representations-using |
| Repo | |
| Framework | |
Improved Modeling of Complex Systems Using Hybrid Physics/Machine Learning/Stochastic Models
| Title | Improved Modeling of Complex Systems Using Hybrid Physics/Machine Learning/Stochastic Models |
| Authors | Anand Ramakrishnan, Warren B. Jackson, Kent Evans |
| Abstract | Combining domain knowledge models with neural models has been challenging. End-to-end trained neural models often perform better (lower Mean Square Error) than domain knowledge models or domain/neural combinations, and the combination is inefficient to train. In this paper, we demonstrate that by composing domain models with machine learning models, by using extrapolative testing sets, and invoking decorrelation objective functions, we create models which can predict more complex systems. The models are interpretable, extrapolative, data-efficient, and capture predictable but complex non-stochastic behavior such as unmodeled degrees of freedom and systemic measurement noise. We apply this improved modeling paradigm to several simulated systems and an actual physical system in the context of system identification. Several ways of composing domain models with neural models are examined for time series, boosting, bagging, and auto-encoding on various systems of varying complexity and non-linearity. Although this work is preliminary, we show that the ability to combine models is a very promising direction for neural modeling. |
| Tasks | Time Series |
| Published | 2020-01-01 |
| URL | https://openreview.net/forum?id=rygfC0VKPS |
| https://openreview.net/pdf?id=rygfC0VKPS | |
| PWC | https://paperswithcode.com/paper/improved-modeling-of-complex-systems-using |
| Repo | |
| Framework | |
Certified Robustness to Adversarial Label-Flipping Attacks via Randomized Smoothing
| Title | Certified Robustness to Adversarial Label-Flipping Attacks via Randomized Smoothing |
| Authors | Anonymous |
| Abstract | This paper considers label-flipping attacks, a type of data poisoning attack where an adversary relabels a small number of examples in a training set in order to degrade the performance of the resulting classifier. In this work, we propose a strategy to build classifiers that are certifiably robust against a strong variant of label-flipping, where the adversary can target each test example independently. In other words, for each test point, our classifier makes a prediction and includes a certification that its prediction would be the same had some number of training labels been changed adversarially. Our approach leverages randomized smoothing, a technique that has previously been used to guarantee test-time robustness to adversarial manipulation of the input to a classifier. Further, we obtain these certified bounds with no additional runtime cost over standard classification. On the Dogfish binary classification task from ImageNet, in the face of an adversary who is allowed to flip 10 labels to individually target each test point, the baseline undefended classifier achieves no more than 29.3% accuracy; we obtain a classifier that maintains 64.2% certified accuracy against the same adversary. |
| Tasks | data poisoning |
| Published | 2020-01-01 |
| URL | https://openreview.net/forum?id=rkgiURVFDS |
| https://openreview.net/pdf?id=rkgiURVFDS | |
| PWC | https://paperswithcode.com/paper/certified-robustness-to-adversarial-label |
| Repo | |
| Framework | |
Exploration via Flow-Based Intrinsic Rewards
| Title | Exploration via Flow-Based Intrinsic Rewards |
| Authors | Anonymous |
| Abstract | Exploration bonuses derived from the novelty of observations in an environment have become a popular approach to motivate exploration for reinforcement learning (RL) agents in the past few years. Recent methods such as curiosity-driven exploration usually estimate the novelty of new observations by the prediction errors of their system dynamics models. In this paper, we introduce the concept of optical flow estimation from the field of computer vision to the RL domain and utilize the errors from optical flow estimation to evaluate the novelty of new observations. We introduce a flow-based intrinsic curiosity module (FICM) capable of learning the motion features and understanding the observations in a more comprehensive and efficient fashion. We evaluate our method and compare it with a number of baselines on several benchmark environments, including Atari games, Super Mario Bros., and ViZDoom. Our results show that the proposed method is superior to the baselines in certain environments, especially for those featuring sophisticated moving patterns or with high-dimensional observation spaces. |
| Tasks | Atari Games, Optical Flow Estimation |
| Published | 2020-01-01 |
| URL | https://openreview.net/forum?id=SkxzSgStPS |
| https://openreview.net/pdf?id=SkxzSgStPS | |
| PWC | https://paperswithcode.com/paper/exploration-via-flow-based-intrinsic-rewards-1 |
| Repo | |
| Framework | |
Robust Domain Randomization for Reinforcement Learning
| Title | Robust Domain Randomization for Reinforcement Learning |
| Authors | Anonymous |
| Abstract | Producing agents that can generalize to a wide range of environments is a significant challenge in reinforcement learning. One method for overcoming this issue is domain randomization, whereby at the start of each training episode some parameters of the environment are randomized so that the agent is exposed to many possible variations. However, domain randomization is highly inefficient and may lead to policies with high variance across domains. In this work, we formalize the domain randomization problem, and show that minimizing the policy’s Lipschitz constant with respect to the randomization parameters leads to low variance in the learned policies. We propose a method where the agent only needs to be trained on one variation of the environment, and its learned state representations are regularized during training to minimize this constant. We conduct experiments that demonstrate that our technique leads to more efficient and robust learning than standard domain randomization, while achieving equal generalization scores. |
| Tasks | |
| Published | 2020-01-01 |
| URL | https://openreview.net/forum?id=H1xSOTVtvH |
| https://openreview.net/pdf?id=H1xSOTVtvH | |
| PWC | https://paperswithcode.com/paper/robust-domain-randomization-for-reinforcement-1 |
| Repo | |
| Framework | |
HyperEmbed: Tradeoffs Between Resources and Performance in NLP Tasks with Hyperdimensional Computing enabled embedding of n-gram statistics
| Title | HyperEmbed: Tradeoffs Between Resources and Performance in NLP Tasks with Hyperdimensional Computing enabled embedding of n-gram statistics |
| Authors | Anonymous |
| Abstract | Recent advances in Deep Learning have led to a significant performance increase on several NLP tasks, however, the models become more and more computationally demanding. Therefore, this paper tackles the domain of computationally efficient algorithms for NLP tasks. In particular, it investigates distributed representations of n-gram statistics of texts. The representations are formed using hyperdimensional computing enabled embedding. These representations then serve as features, which are used as input to standard classifiers. We investigate the applicability of the embedding on one large and three small standard datasets for classification tasks using nine classifiers. The embedding achieved on par F1 scores while decreasing the time and memory requirements by several times compared to the conventional n-gram statistics, e.g., for one of the classifiers on a small dataset, the memory reduction was 6.18 times; while train and test speed-ups were 4.62 and 3.84 times, respectively. For many classifiers on the large dataset, the memory reduction was about 100 times and train and test speed-ups were over 100 times. More importantly, the usage of distributed representations formed via hyperdimensional computing allows dissecting the strict dependency between the dimensionality of the representation and the parameters of n-gram statistics, thus, opening a room for tradeoffs. |
| Tasks | |
| Published | 2020-01-01 |
| URL | https://openreview.net/forum?id=SJgzXaNFwS |
| https://openreview.net/pdf?id=SJgzXaNFwS | |
| PWC | https://paperswithcode.com/paper/hyperembed-tradeoffs-between-resources-and |
| Repo | |
| Framework | |
Hydra: Preserving Ensemble Diversity for Model Distillation
| Title | Hydra: Preserving Ensemble Diversity for Model Distillation |
| Authors | Anonymous |
| Abstract | Ensembles of models have been empirically shown to improve predictive performance and to yield robust measures of uncertainty. However, they are expensive in computation and memory. Therefore, recent research has focused on distilling ensembles into a single compact model, reducing the computational and memory burden of the ensemble while trying to preserve its predictive behavior. Most existing distillation formulations summarize the ensemble by capturing its average predictions. As a result, the diversity of the ensemble predictions, stemming from each individual member, is lost. Thus the distilled model cannot provide a measure of uncertainty comparable to that of the original ensemble. To retain more faithfully the diversity of the ensemble, we propose a distillation method based on a single multi-headed neural network, which we refer to as Hydra. The shared body network learns a joint feature representation that enables each head to capture the predictive behavior of each ensemble member. We demonstrate that with a slight increase in parameter count, Hydra improves distillation performance on classification and regression settings while capturing the uncertainty behaviour of the original ensemble over both in-domain and out-of-distribution tasks. |
| Tasks | |
| Published | 2020-01-01 |
| URL | https://openreview.net/forum?id=ByeaXeBFvH |
| https://openreview.net/pdf?id=ByeaXeBFvH | |
| PWC | https://paperswithcode.com/paper/hydra-preserving-ensemble-diversity-for-model |
| Repo | |
| Framework | |
Are there any ‘object detectors’ in the hidden layers of CNNs trained to identify objects or scenes?
| Title | Are there any ‘object detectors’ in the hidden layers of CNNs trained to identify objects or scenes? |
| Authors | Anonymous |
| Abstract | Various methods of measuring unit selectivity have been developed with the aim of better understanding how neural networks work. But the different measures provide divergent estimates of selectivity, and this has led to different conclusions regarding the conditions in which selective object representations are learned and the functional relevance of these representations. In an attempt to better characterize object selectivity, we undertake a comparison of various selectivity measures on a large set of units in AlexNet, including localist selectivity, precision, class-conditional mean activity selectivity (CCMAS), network dissection, the human interpretation of activation maximization (AM) images, and standard signal-detection measures. We find that the different measures provide different estimates of object selectivity, with precision and CCMAS measures providing misleadingly high estimates. Indeed, the most selective units had a poor hit-rate or a high false-alarm rate (or both) in object classification, making them poor object detectors. We fail to find any units that are even remotely as selective as the ‘grandmother cell’ units reported in recurrent neural networks. In order to generalize these results, we compared selectivity measures on a few units in VGG-16 and GoogLeNet trained on the ImageNet or Places-365 datasets that have been described as ‘object detectors’. Again, we find poor hit-rates and high false-alarm rates for object classification. |
| Tasks | Object Classification |
| Published | 2020-01-01 |
| URL | https://openreview.net/forum?id=Skltqh4KvB |
| https://openreview.net/pdf?id=Skltqh4KvB | |
| PWC | https://paperswithcode.com/paper/are-there-any-object-detectors-in-the-hidden |
| Repo | |
| Framework | |
StructPool: Structured Graph Pooling via Conditional Random Fields
| Title | StructPool: Structured Graph Pooling via Conditional Random Fields |
| Authors | Anonymous |
| Abstract | Learning high-level representations for graphs is of great importance for graph analysis tasks. In addition to graph convolution, graph pooling is an important but less explored research area. In particular, most of existing graph pooling techniques do not consider the graph structural information explicitly. We argue that such information is important and develop a novel graph pooling technique, know as the StructPool, in this work. We consider the graph pooling as a node clustering problem, which requires the learning of a cluster assignment matrix. We propose to formulate it as a structured prediction problem and employ conditional random fields to capture the relationships among assignments of different nodes. We also generalize our method to incorporate graph topological information in designing the Gibbs energy function. Experimental results on multiple datasets demonstrate the effectiveness of our proposed StructPool. |
| Tasks | Structured Prediction |
| Published | 2020-01-01 |
| URL | https://openreview.net/forum?id=BJxg_hVtwH |
| https://openreview.net/pdf?id=BJxg_hVtwH | |
| PWC | https://paperswithcode.com/paper/structpool-structured-graph-pooling-via |
| Repo | |
| Framework | |
Enhancing the Transformer with explicit relational encoding for math problem solving
| Title | Enhancing the Transformer with explicit relational encoding for math problem solving |
| Authors | Anonymous |
| Abstract | We incorporate Tensor-Product Representations within the Transformer in order to better support the explicit representation of relation structure. Our Tensor-Product Transformer (TP-Transformer) sets a new state of the art on the recently-introduced Mathematics Dataset containing 56 categories of free-form math word-problems. The essential component of the model is a novel attention mechanism, called TP-Attention, which explicitly encodes the relations between each Transformer cell and the other cells from which values have been retrieved by attention. TP-Attention goes beyond linear combination of retrieved values, strengthening representation-building and resolving ambiguities introduced by multiple layers of regular attention. The TP-Transformer’s attention maps give better insights into how it is capable of solving the Mathematics Dataset’s challenging problems. Pretrained models and code will be made available after publication. |
| Tasks | |
| Published | 2020-01-01 |
| URL | https://openreview.net/forum?id=B1xfElrKPr |
| https://openreview.net/pdf?id=B1xfElrKPr | |
| PWC | https://paperswithcode.com/paper/enhancing-the-transformer-with-explicit |
| Repo | |
| Framework | |
Style-based Encoder Pre-training for Multi-modal Image Synthesis
| Title | Style-based Encoder Pre-training for Multi-modal Image Synthesis |
| Authors | Anonymous |
| Abstract | Image-to-image (I2I) translation aims to translate images from one domain to another. To tackle the multi-modal version of I2I translation, where input and output domains have a one-to-many relation, an extra latent input is provided to the generator to specify a particular output. Recent works propose involved training objectives to learn a latent embedding, jointly with the generator, that models the distribution of possible outputs. Alternatively, we study a simple, yet powerful pre-training strategy for multi-modal I2I translation. We first pre-train an encoder, using a proxy task, to encode the style of an image, such as color and texture, into a low-dimensional latent style vector. Then we train a generator to transform an input image along with a style-code to the output domain. Our generator achieves state-of-the-art results on several benchmarks with a training objective that includes just a GAN loss and a reconstruction loss, which simplifies and speeds up the training significantly compared to competing approaches. We further study the contribution of different loss terms to learning the task of multi-modal I2I translation, and finally we show that the learned style embedding is not dependent on the target domain and generalizes well to other domains. |
| Tasks | Image Generation |
| Published | 2020-01-01 |
| URL | https://openreview.net/forum?id=rkgrbTNtDr |
| https://openreview.net/pdf?id=rkgrbTNtDr | |
| PWC | https://paperswithcode.com/paper/style-based-encoder-pre-training-for-multi |
| Repo | |
| Framework | |
Augmenting Genetic Algorithms with Deep Neural Networks for Exploring the Chemical Space
| Title | Augmenting Genetic Algorithms with Deep Neural Networks for Exploring the Chemical Space |
| Authors | Anonymous |
| Abstract | Challenges in natural sciences can often be phrased as optimization problems. Machine learning techniques have recently been applied to solve such problems. One example in chemistry is the design of tailor-made organic materials and molecules, which requires efficient methods to explore the chemical space. We present a genetic algorithm (GA) that is enhanced with a neural network (DNN) based discriminator model to improve the diversity of generated molecules and at the same time steer the GA. We show that our algorithm outperforms other generative models in optimization tasks. We furthermore present a way to increase interpretability of genetic algorithms, which helped us to derive design principles |
| Tasks | |
| Published | 2020-01-01 |
| URL | https://openreview.net/forum?id=H1lmyRNFvr |
| https://openreview.net/pdf?id=H1lmyRNFvr | |
| PWC | https://paperswithcode.com/paper/augmenting-genetic-algorithms-with-deep-1 |
| Repo | |
| Framework | |
Dynamically Pruned Message Passing Networks for Large-scale Knowledge Graph Reasoning
| Title | Dynamically Pruned Message Passing Networks for Large-scale Knowledge Graph Reasoning |
| Authors | Anonymous |
| Abstract | We propose Dynamically Pruned Message Passing Networks (DPMPN) for large-scale knowledge graph reasoning. In contrast to existing models, embedding-based or path-based, we learn an input-dependent subgraph to explicitly model a sequential reasoning process. Each subgraph is dynamically constructed, expanding itself selectively under a flow-style attention mechanism. In this way, we can not only construct graphical explanations to interpret prediction, but also prune message passing in Graph Neural Networks (GNNs) to scale with the size of graphs. We take the inspiration from the consciousness prior proposed by Bengio to design a two-GNN framework to encode global input-invariant graph-structured representation and learn local input-dependent one coordinated by an attention module. Experiments show the reasoning capability in our model that is providing a clear graphical explanation as well as predicting results accurately, outperforming most state-of-the-art methods in knowledge base completion tasks. |
| Tasks | Knowledge Base Completion |
| Published | 2020-01-01 |
| URL | https://openreview.net/forum?id=rkeuAhVKvB |
| https://openreview.net/pdf?id=rkeuAhVKvB | |
| PWC | https://paperswithcode.com/paper/dynamically-pruned-message-passing-networks-1 |
| Repo | |
| Framework | |
Trajectory representation learning for Multi-Task NMRDPs planning
| Title | Trajectory representation learning for Multi-Task NMRDPs planning |
| Authors | Anonymous |
| Abstract | Expanding Non Markovian Reward Decision Processes (NMRDP) into Markov Decision Processes (MDP) enables the use of state of the art Reinforcement Learning (RL) techniques to identify optimal policies. In this paper an approach to exploring NMRDPs and expanding them into MDPs, without the prior knowledge of the reward structure, is proposed. The non Markovianity of the reward function is disentangled under the assumption that sets of similar and dissimilar trajectory batches can be sampled. More precisely, within the same batch, measuring the similarity between any couple of trajectories is permitted, although comparing trajectories from different batches is not possible. A modified version of the triplet loss is optimised to construct a representation of the trajectories under which rewards become Markovian. |
| Tasks | Representation Learning |
| Published | 2020-01-01 |
| URL | https://openreview.net/forum?id=rkeYvaNKPr |
| https://openreview.net/pdf?id=rkeYvaNKPr | |
| PWC | https://paperswithcode.com/paper/trajectory-representation-learning-for-multi |
| Repo | |
| Framework | |
