Konrad Zolna

Scott Reed, Konrad Zolna, Emilio Parisotto et al.

Inspired by progress in large-scale language modeling, we apply a similar approach towards building a single generalist agent beyond the realm of text outputs. The agent, which we refer to as Gato, works as a multi-modal, multi-task, multi-embodiment generalist policy. The same network with the same weights can play Atari, caption images, chat, stack blocks with a real robot arm and much more, deciding based on its context whether to output text, joint torques, button presses, or other tokens. In this report we describe the model and the data, and document the current capabilities of Gato.

Imagen-Team-Google, Jason Baldridge, Jakob Bauer et al.

We introduce Imagen 3, a latent diffusion model that generates high quality images from text prompts. We describe our quality and responsibility evaluations. Imagen 3 is preferred over other state-of-the-art (SOTA) models at the time of evaluation. In addition, we discuss issues around safety and representation, as well as methods we used to minimize the potential harm of our models.

Jake Bruce, Michael Dennis, Ashley Edwards et al. · oxford

We introduce Genie, the first generative interactive environment trained in an unsupervised manner from unlabelled Internet videos. The model can be prompted to generate an endless variety of action-controllable virtual worlds described through text, synthetic images, photographs, and even sketches. At 11B parameters, Genie can be considered a foundation world model. It is comprised of a spatiotemporal video tokenizer, an autoregressive dynamics model, and a simple and scalable latent action model. Genie enables users to act in the generated environments on a frame-by-frame basis despite training without any ground-truth action labels or other domain-specific requirements typically found in the world model literature. Further the resulting learned latent action space facilitates training agents to imitate behaviors from unseen videos, opening the path for training generalist agents of the future.

Caglar Gulcehre, Ziyu Wang, Alexander Novikov et al.

Offline methods for reinforcement learning have a potential to help bridge the gap between reinforcement learning research and real-world applications. They make it possible to learn policies from offline datasets, thus overcoming concerns associated with online data collection in the real-world, including cost, safety, or ethical concerns. In this paper, we propose a benchmark called RL Unplugged to evaluate and compare offline RL methods. RL Unplugged includes data from a diverse range of domains including games (e.g., Atari benchmark) and simulated motor control problems (e.g., DM Control Suite). The datasets include domains that are partially or fully observable, use continuous or discrete actions, and have stochastic vs. deterministic dynamics. We propose detailed evaluation protocols for each domain in RL Unplugged and provide an extensive analysis of supervised learning and offline RL methods using these protocols. We will release data for all our tasks and open-source all algorithms presented in this paper. We hope that our suite of benchmarks will increase the reproducibility of experiments and make it possible to study challenging tasks with a limited computational budget, thus making RL research both more systematic and more accessible across the community. Moving forward, we view RL Unplugged as a living benchmark suite that will evolve and grow with datasets contributed by the research community and ourselves. Our project page is available on https://git.io/JJUhd.

Konrad Zolna, Michal Zajac, Negar Rostamzadeh et al.

Neural networks are prone to adversarial attacks. In general, such attacks deteriorate the quality of the input by either slightly modifying most of its pixels, or by occluding it with a patch. In this paper, we propose a method that keeps the image unchanged and only adds an adversarial framing on the border of the image. We show empirically that our method is able to successfully attack state-of-the-art methods on both image and video classification problems. Notably, the proposed method results in a universal attack which is very fast at test time. Source code can be found at https://github.com/zajaczajac/adv_framing .

Konrad Zolna, Krzysztof J. Geras, Kyunghyun Cho

Currently available methods for extracting saliency maps identify parts of the input which are the most important to a specific fixed classifier. We show that this strong dependence on a given classifier hinders their performance. To address this problem, we propose classifier-agnostic saliency map extraction, which finds all parts of the image that any classifier could use, not just one given in advance. We observe that the proposed approach extracts higher quality saliency maps than prior work while being conceptually simple and easy to implement. The method sets the new state of the art result for localization task on the ImageNet data, outperforming all existing weakly-supervised localization techniques, despite not using the ground truth labels at the inference time. The code reproducing the results is available at https://github.com/kondiz/casme . The final version of this manuscript is published in Computer Vision and Image Understanding and is available online at https://doi.org/10.1016/j.cviu.2020.102969 .

Konrad Zolna, Serkan Cabi, Yutian Chen et al.

As the AI community increasingly adopts large-scale models, it is crucial to develop general and flexible tools to integrate them. We introduce Gather-Attend-Scatter (GATS), a novel module that enables seamless combination of pretrained foundation models, both trainable and frozen, into larger multimodal networks. GATS empowers AI systems to process and generate information across multiple modalities at different rates. In contrast to traditional fine-tuning, GATS allows for the original component models to remain frozen, avoiding the risk of them losing important knowledge acquired during the pretraining phase. We demonstrate the utility and versatility of GATS with a few experiments across games, robotics, and multimodal input-output systems.

Jakub Chłędowski, Adam Polak, Bartosz Szabucki et al.

Effective caching is crucial for the performance of modern-day computing systems. A key optimization problem arising in caching -- which item to evict to make room for a new item -- cannot be optimally solved without knowing the future. There are many classical approximation algorithms for this problem, but more recently researchers started to successfully apply machine learning to decide what to evict by discovering implicit input patterns and predicting the future. While machine learning typically does not provide any worst-case guarantees, the new field of learning-augmented algorithms proposes solutions that leverage classical online caching algorithms to make the machine-learned predictors robust. We are the first to comprehensively evaluate these learning-augmented algorithms on real-world caching datasets and state-of-the-art machine-learned predictors. We show that a straightforward method -- blindly following either a predictor or a classical robust algorithm, and switching whenever one becomes worse than the other -- has only a low overhead over a well-performing predictor, while competing with classical methods when the coupled predictor fails, thus providing a cheap worst-case insurance.

Caglar Gulcehre, Sergio Gómez Colmenarejo, Ziyu Wang et al.

Offline reinforcement learning restricts the learning process to rely only on logged-data without access to an environment. While this enables real-world applications, it also poses unique challenges. One important challenge is dealing with errors caused by the overestimation of values for state-action pairs not well-covered by the training data. Due to bootstrapping, these errors get amplified during training and can lead to divergence, thereby crippling learning. To overcome this challenge, we introduce Regularized Behavior Value Estimation (R-BVE). Unlike most approaches, which use policy improvement during training, R-BVE estimates the value of the behavior policy during training and only performs policy improvement at deployment time. Further, R-BVE uses a ranking regularisation term that favours actions in the dataset that lead to successful outcomes. We provide ample empirical evidence of R-BVE's effectiveness, including state-of-the-art performance on the RL Unplugged ATARI dataset. We also test R-BVE on new datasets, from bsuite and a challenging DeepMind Lab task, and show that R-BVE outperforms other state-of-the-art discrete control offline RL methods.

Ksenia Konyushkova, Konrad Zolna, Yusuf Aytar et al.

In offline reinforcement learning (RL) agents are trained using a logged dataset. It appears to be the most natural route to attack real-life applications because in domains such as healthcare and robotics interactions with the environment are either expensive or unethical. Training agents usually requires reward functions, but unfortunately, rewards are seldom available in practice and their engineering is challenging and laborious. To overcome this, we investigate reward learning under the constraint of minimizing human reward annotations. We consider two types of supervision: timestep annotations and demonstrations. We propose semi-supervised learning algorithms that learn from limited annotations and incorporate unlabelled data. In our experiments with a simulated robotic arm, we greatly improve upon behavioural cloning and closely approach the performance achieved with ground truth rewards. We further investigate the relationship between the quality of the reward model and the final policies. We notice, for example, that the reward models do not need to be perfect to result in useful policies.

Konrad Zolna, Alexander Novikov, Ksenia Konyushkova et al.

Behavior cloning (BC) is often practical for robot learning because it allows a policy to be trained offline without rewards, by supervised learning on expert demonstrations. However, BC does not effectively leverage what we will refer to as unlabeled experience: data of mixed and unknown quality without reward annotations. This unlabeled data can be generated by a variety of sources such as human teleoperation, scripted policies and other agents on the same robot. Towards data-driven offline robot learning that can use this unlabeled experience, we introduce Offline Reinforced Imitation Learning (ORIL). ORIL first learns a reward function by contrasting observations from demonstrator and unlabeled trajectories, then annotates all data with the learned reward, and finally trains an agent via offline reinforcement learning. Across a diverse set of continuous control and simulated robotic manipulation tasks, we show that ORIL consistently outperforms comparable BC agents by effectively leveraging unlabeled experience.

Tom Le Paine, Cosmin Paduraru, Andrea Michi et al.

Offline reinforcement learning (RL purely from logged data) is an important avenue for deploying RL techniques in real-world scenarios. However, existing hyperparameter selection methods for offline RL break the offline assumption by evaluating policies corresponding to each hyperparameter setting in the environment. This online execution is often infeasible and hence undermines the main aim of offline RL. Therefore, in this work, we focus on \textit{offline hyperparameter selection}, i.e. methods for choosing the best policy from a set of many policies trained using different hyperparameters, given only logged data. Through large-scale empirical evaluation we show that: 1) offline RL algorithms are not robust to hyperparameter choices, 2) factors such as the offline RL algorithm and method for estimating Q values can have a big impact on hyperparameter selection, and 3) when we control those factors carefully, we can reliably rank policies across hyperparameter choices, and therefore choose policies which are close to the best policy in the set. Overall, our results present an optimistic view that offline hyperparameter selection is within reach, even in challenging tasks with pixel observations, high dimensional action spaces, and long horizon.

Ziyu Wang, Alexander Novikov, Konrad Zolna et al.

Offline reinforcement learning (RL), also known as batch RL, offers the prospect of policy optimization from large pre-recorded datasets without online environment interaction. It addresses challenges with regard to the cost of data collection and safety, both of which are particularly pertinent to real-world applications of RL. Unfortunately, most off-policy algorithms perform poorly when learning from a fixed dataset. In this paper, we propose a novel offline RL algorithm to learn policies from data using a form of critic-regularized regression (CRR). We find that CRR performs surprisingly well and scales to tasks with high-dimensional state and action spaces -- outperforming several state-of-the-art offline RL algorithms by a significant margin on a wide range of benchmark tasks.

Konrad Zolna, Chitwan Saharia, Leonard Boussioux et al.

In adversarial imitation learning, a discriminator is trained to differentiate agent episodes from expert demonstrations representing the desired behavior. However, as the trained policy learns to be more successful, the negative examples (the ones produced by the agent) become increasingly similar to expert ones. Despite the fact that the task is successfully accomplished in some of the agent's trajectories, the discriminator is trained to output low values for them. We hypothesize that this inconsistent training signal for the discriminator can impede its learning, and consequently leads to worse overall performance of the agent. We show experimental evidence for this hypothesis and that the 'False Negatives' (i.e. successful agent episodes) significantly hinder adversarial imitation learning, which is the first contribution of this paper. Then, we propose a method to alleviate the impact of false negatives and test it on the BabyAI environment. This method consistently improves sample efficiency over the baselines by at least an order of magnitude.

Konrad Zolna, Scott Reed, Alexander Novikov et al.

We show that a critical vulnerability in adversarial imitation is the tendency of discriminator networks to learn spurious associations between visual features and expert labels. When the discriminator focuses on task-irrelevant features, it does not provide an informative reward signal, leading to poor task performance. We analyze this problem in detail and propose a solution that outperforms standard Generative Adversarial Imitation Learning (GAIL). Our proposed method, Task-Relevant Adversarial Imitation Learning (TRAIL), uses constrained discriminator optimization to learn informative rewards. In comprehensive experiments, we show that TRAIL can solve challenging robotic manipulation tasks from pixels by imitating human operators without access to any task rewards, and clearly outperforms comparable baseline imitation agents, including those trained via behaviour cloning and conventional GAIL.

Serkan Cabi, Sergio Gómez Colmenarejo, Alexander Novikov et al.

We present a framework for data-driven robotics that makes use of a large dataset of recorded robot experience and scales to several tasks using learned reward functions. We show how to apply this framework to accomplish three different object manipulation tasks on a real robot platform. Given demonstrations of a task together with task-agnostic recorded experience, we use a special form of human annotation as supervision to learn a reward function, which enables us to deal with real-world tasks where the reward signal cannot be acquired directly. Learned rewards are used in combination with a large dataset of experience from different tasks to learn a robot policy offline using batch RL. We show that using our approach it is possible to train agents to perform a variety of challenging manipulation tasks including stacking rigid objects and handling cloth.

Konrad Zolna, Thibault Asselborn, Caroline Jolly et al.

Handwriting disorder (termed dysgraphia) is a far from a singular problem as nearly 8.6% of the population in France is considered dysgraphic. Moreover, research highlights the fundamental importance to detect and remediate these handwriting difficulties as soon as possible as they may affect a child's entire life, undermining performance and self-confidence in a wide variety of school activities. At the moment, the detection of handwriting difficulties is performed through a standard test called BHK. This detection, performed by therapists, is laborious because of its high cost and subjectivity. We present a digital approach to identify and characterize handwriting difficulties via a Recurrent Neural Network model (RNN). The child under investigation is asked to write on a graphics tablet all the letters of the alphabet as well as the ten digits. Once complete, the RNN delivers a diagnosis in a few milliseconds and demonstrates remarkable efficiency as it correctly identifies more than 90% of children diagnosed as dysgraphic using the BHK test. The main advantage of our tablet-based system is that it captures the dynamic features of writing -- something a human expert, such as a teacher, is unable to do. We show that incorporating the dynamic information available by the use of tablet is highly beneficial to our digital test to discriminate between typically-developing and dysgraphic children.

Michał Zając, Konrad Zolna, Stanisław Jastrzębski

Recent work has shown that using unlabeled data in semi-supervised learning is not always beneficial and can even hurt generalization, especially when there is a class mismatch between the unlabeled and labeled examples. We investigate this phenomenon for image classification on the CIFAR-10 and the ImageNet datasets, and with many other forms of domain shifts applied (e.g. salt-and-pepper noise). Our main contribution is Split Batch Normalization (Split-BN), a technique to improve SSL when the additional unlabeled data comes from a shifted distribution. We achieve it by using separate batch normalization statistics for unlabeled examples. Due to its simplicity, we recommend it as a standard practice. Finally, we analyse how domain shift affects the SSL training process. In particular, we find that during training the statistics of hidden activations in late layers become markedly different between the unlabeled and the labeled examples.

Konrad Zolna, Negar Rostamzadeh, Yoshua Bengio et al.

Imitation learning is an effective alternative approach to learn a policy when the reward function is sparse. In this paper, we consider a challenging setting where an agent and an expert use different actions from each other. We assume that the agent has access to a sparse reward function and state-only expert observations. We propose a method which gradually balances between the imitation learning cost and the reinforcement learning objective. In addition, this method adapts the agent's policy based on either mimicking expert behavior or maximizing sparse reward. We show, through navigation scenarios, that (i) an agent is able to efficiently leverage sparse rewards to outperform standard state-only imitation learning, (ii) it can learn a policy even when its actions are different from the expert, and (iii) the performance of the agent is not bounded by that of the expert, due to the optimized usage of sparse rewards.

Nan Rosemary Ke, Konrad Zolna, Alessandro Sordoni et al.

Recurrent Neural Networks (RNNs) with attention mechanisms have obtained state-of-the-art results for many sequence processing tasks. Most of these models use a simple form of encoder with attention that looks over the entire sequence and assigns a weight to each token independently. We present a mechanism for focusing RNN encoders for sequence modelling tasks which allows them to attend to key parts of the input as needed. We formulate this using a multi-layer conditional sequence encoder that reads in one token at a time and makes a discrete decision on whether the token is relevant to the context or question being asked. The discrete gating mechanism takes in the context embedding and the current hidden state as inputs and controls information flow into the layer above. We train it using policy gradient methods. We evaluate this method on several types of tasks with different attributes. First, we evaluate the method on synthetic tasks which allow us to evaluate the model for its generalization ability and probe the behavior of the gates in more controlled settings. We then evaluate this approach on large scale Question Answering tasks including the challenging MS MARCO and SearchQA tasks. Our models shows consistent improvements for both tasks over prior work and our baselines. It has also shown to generalize significantly better on synthetic tasks as compared to the baselines.

Konrad Zolna, Devansh Arpit, Dendi Suhubdy et al.

Recurrent neural networks (RNNs) are important class of architectures among neural networks useful for language modeling and sequential prediction. However, optimizing RNNs is known to be harder compared to feed-forward neural networks. A number of techniques have been proposed in literature to address this problem. In this paper we propose a simple technique called fraternal dropout that takes advantage of dropout to achieve this goal. Specifically, we propose to train two identical copies of an RNN (that share parameters) with different dropout masks while minimizing the difference between their (pre-softmax) predictions. In this way our regularization encourages the representations of RNNs to be invariant to dropout mask, thus being robust. We show that our regularization term is upper bounded by the expectation-linear dropout objective which has been shown to address the gap due to the difference between the train and inference phases of dropout. We evaluate our model and achieve state-of-the-art results in sequence modeling tasks on two benchmark datasets - Penn Treebank and Wikitext-2. We also show that our approach leads to performance improvement by a significant margin in image captioning (Microsoft COCO) and semi-supervised (CIFAR-10) tasks.

Konrad Zolna

The method presented extends a given regression neural network to make its performance improve. The modification affects the learning procedure only, hence the extension may be easily omitted during evaluation without any change in prediction. It means that the modified model may be evaluated as quickly as the original one but tends to perform better. This improvement is possible because the modification gives better expressive power, provides better behaved gradients and works as a regularization. The knowledge gained by the temporarily extended neural network is contained in the parameters shared with the original neural network. The only cost is an increase in learning time.