Siqi Liu

Ian Gemp, Thomas Anthony, Yoram Bachrach et al. · deepmind

The Game Theory & Multi-Agent team at DeepMind studies several aspects of multi-agent learning ranging from computing approximations to fundamental concepts in game theory to simulating social dilemmas in rich spatial environments and training 3-d humanoids in difficult team coordination tasks. A signature aim of our group is to use the resources and expertise made available to us at DeepMind in deep reinforcement learning to explore multi-agent systems in complex environments and use these benchmarks to advance our understanding. Here, we summarise the recent work of our team and present a taxonomy that we feel highlights many important open challenges in multi-agent research.

Xinyi Yu, Guanbin Li, Wei Lou et al.

Nuclei segmentation is a fundamental but challenging task in the quantitative analysis of histopathology images. Although fully-supervised deep learning-based methods have made significant progress, a large number of labeled images are required to achieve great segmentation performance. Considering that manually labeling all nuclei instances for a dataset is inefficient, obtaining a large-scale human-annotated dataset is time-consuming and labor-intensive. Therefore, augmenting a dataset with only a few labeled images to improve the segmentation performance is of significant research and application value. In this paper, we introduce the first diffusion-based augmentation method for nuclei segmentation. The idea is to synthesize a large number of labeled images to facilitate training the segmentation model. To achieve this, we propose a two-step strategy. In the first step, we train an unconditional diffusion model to synthesize the Nuclei Structure that is defined as the representation of pixel-level semantic and distance transform. Each synthetic nuclei structure will serve as a constraint on histopathology image synthesis and is further post-processed to be an instance map. In the second step, we train a conditioned diffusion model to synthesize histopathology images based on nuclei structures. The synthetic histopathology images paired with synthetic instance maps will be added to the real dataset for training the segmentation model. The experimental results show that by augmenting 10% labeled real dataset with synthetic samples, one can achieve comparable segmentation results with the fully-supervised baseline. The code is released in: https://github.com/lhaof/Nudiff

Bobak Shahriari, Abbas Abdolmaleki, Arunkumar Byravan et al.

Actor-critic algorithms that make use of distributional policy evaluation have frequently been shown to outperform their non-distributional counterparts on many challenging control tasks. Examples of this behavior include the D4PG and DMPO algorithms as compared to DDPG and MPO, respectively [Barth-Maron et al., 2018; Hoffman et al., 2020]. However, both agents rely on the C51 critic for value estimation.One major drawback of the C51 approach is its requirement of prior knowledge about the minimum andmaximum values a policy can attain as well as the number of bins used, which fixes the resolution ofthe distributional estimate. While the DeepMind control suite of tasks utilizes standardized rewards and episode lengths, thus enabling the entire suite to be solved with a single setting of these hyperparameters, this is often not the case. This paper revisits a natural alternative that removes this requirement, namelya mixture of Gaussians, and a simple sample-based loss function to train it in an off-policy regime. We empirically evaluate its performance on a broad range of continuous control tasks and demonstrate that it eliminates the need for these distributional hyperparameters and achieves state-of-the-art performance on a variety of challenging tasks (e.g. the humanoid, dog, quadruped, and manipulator domains). Finallywe provide an implementation in the Acme agent repository.

Luke Marris, Ian Gemp, Thomas Anthony et al.

Solution concepts such as Nash Equilibria, Correlated Equilibria, and Coarse Correlated Equilibria are useful components for many multiagent machine learning algorithms. Unfortunately, solving a normal-form game could take prohibitive or non-deterministic time to converge, and could fail. We introduce the Neural Equilibrium Solver which utilizes a special equivariant neural network architecture to approximately solve the space of all games of fixed shape, buying speed and determinism. We define a flexible equilibrium selection framework, that is capable of uniquely selecting an equilibrium that minimizes relative entropy, or maximizes welfare. The network is trained without needing to generate any supervised training data. We show remarkable zero-shot generalization to larger games. We argue that such a network is a powerful component for many possible multiagent algorithms.

Wei Lou, Xinyi Yu, Chenyu Liu et al.

Cell segmentation for multi-modal microscopy images remains a challenge due to the complex textures, patterns, and cell shapes in these images. To tackle the problem, we first develop an automatic cell classification pipeline to label the microscopy images based on their low-level image characteristics, and then train a classification model based on the category labels. Afterward, we train a separate segmentation model for each category using the images in the corresponding category. Besides, we further deploy two types of segmentation models to segment cells with roundish and irregular shapes respectively. Moreover, an efficient and powerful backbone model is utilized to enhance the efficiency of our segmentation model. Evaluated on the Tuning Set of NeurIPS 2022 Cell Segmentation Challenge, our method achieves an F1-score of 0.8795 and the running time for all cases is within the time tolerance.

Yong-Lu Li, Yue Xu, Xinyu Xu et al.

Understanding objects is a central building block of artificial intelligence, especially for embodied AI. Even though object recognition excels with deep learning, current machines still struggle to learn higher-level knowledge, e.g., what attributes an object has, and what can we do with an object. In this work, we propose a challenging Object Concept Learning (OCL) task to push the envelope of object understanding. It requires machines to reason out object affordances and simultaneously give the reason: what attributes make an object possesses these affordances. To support OCL, we build a densely annotated knowledge base including extensive labels for three levels of object concept (category, attribute, affordance), and the causal relations of three levels. By analyzing the causal structure of OCL, we present a baseline, Object Concept Reasoning Network (OCRN). It leverages causal intervention and concept instantiation to infer the three levels following their causal relations. In experiments, OCRN effectively infers the object knowledge while following the causalities well. Our data and code are available at https://mvig-rhos.com/ocl.

Siqi Liu, Andreas Lehrmann

Deep learning has shown impressive results in a variety of time series forecasting tasks, where modeling the conditional distribution of the future given the past is the essence. However, when this conditional distribution is non-stationary, it poses challenges for these models to learn consistently and to predict accurately. In this work, we propose a new method to model non-stationary conditional distributions over time by clearly decoupling stationary conditional distribution modeling from non-stationary dynamics modeling. Our method is based on a Bayesian dynamic model that can adapt to conditional distribution changes and a deep conditional distribution model that handles multivariate time series using a factorized output space. Our experimental results on synthetic and real-world datasets show that our model can adapt to non-stationary time series better than state-of-the-art deep learning solutions.

Haiyang Wang, Guangyu Zhou, Siqi Liu et al.

Motivation: Predicting Drug-Target Interaction (DTI) is a well-studied topic in bioinformatics due to its relevance in the fields of proteomics and pharmaceutical research. Although many machine learning methods have been successfully applied in this task, few of them aim at leveraging the inherent heterogeneous graph structure in the DTI network to address the challenge. For better learning and interpreting the DTI topological structure and the similarity, it is desirable to have methods specifically for predicting interactions from the graph structure. Results: We present an end-to-end framework, DTI-GAT (Drug-Target Interaction prediction with Graph Attention networks) for DTI predictions. DTI-GAT incorporates a deep neural network architecture that operates on graph-structured data with the attention mechanism, which leverages both the interaction patterns and the features of drug and protein sequences. DTI-GAT facilitates the interpretation of the DTI topological structure by assigning different attention weights to each node with the self-attention mechanism. Experimental evaluations show that DTI-GAT outperforms various state-of-the-art systems on the binary DTI prediction problem. Moreover, the independent study results further demonstrate that our model can be generalized better than other conventional methods. Availability: The source code and all datasets are available at https://github.com/Haiyang-W/DTI-GRAPH

Yuval Tassa, Saran Tunyasuvunakool, Alistair Muldal et al.

The dm_control software package is a collection of Python libraries and task suites for reinforcement learning agents in an articulated-body simulation. A MuJoCo wrapper provides convenient bindings to functions and data structures. The PyMJCF and Composer libraries enable procedural model manipulation and task authoring. The Control Suite is a fixed set of tasks with standardised structure, intended to serve as performance benchmarks. The Locomotion framework provides high-level abstractions and examples of locomotion tasks. A set of configurable manipulation tasks with a robot arm and snap-together bricks is also included. dm_control is publicly available at https://www.github.com/deepmind/dm_control

George Shaikovski, Adam Casson, Kristen Severson et al.

Foundation models in computational pathology promise to unlock the development of new clinical decision support systems and models for precision medicine. However, there is a mismatch between most clinical analysis, which is defined at the level of one or more whole slide images, and foundation models to date, which process the thousands of image tiles contained in a whole slide image separately. The requirement to train a network to aggregate information across a large number of tiles in multiple whole slide images limits these models' impact. In this work, we present a slide-level foundation model for H&E-stained histopathology, PRISM, that builds on Virchow tile embeddings and leverages clinical report text for pre-training. Using the tile embeddings, PRISM produces slide-level embeddings with the ability to generate clinical reports, resulting in several modes of use. Using text prompts, PRISM achieves zero-shot cancer detection and sub-typing performance approaching and surpassing that of a supervised aggregator model. Using the slide embeddings with linear classifiers, PRISM surpasses supervised aggregator models. Furthermore, we demonstrate that fine-tuning of the PRISM slide encoder yields label-efficient training for biomarker prediction, a task that typically suffers from low availability of training data; an aggregator initialized with PRISM and trained on as little as 10% of the training data can outperform a supervised baseline that uses all of the data.

Somjit Nath, Yik Chau Lui, Siqi Liu

Event sequence data record the occurrences of events in continuous time. Event sequence forecasting based on temporal point processes (TPPs) has been extensively studied, but outlier or anomaly detection, especially without any supervision from humans, is still underexplored. In this work, we develop, to the best our knowledge, the first unsupervised outlier detection approach to detecting abnormal events. Our novel unsupervised outlier detection framework is based on ideas from generative adversarial networks (GANs) and reinforcement learning (RL). We train a 'generator' that corrects outliers in the data with a 'discriminator' that learns to discriminate the corrected data from the real data, which may contain outliers. A key insight is that if the generator made a mistake in the correction, it would generate anomalies that are different from the anomalies in the real data, so it serves as data augmentation for the discriminator learning. Different from typical GAN-based outlier detection approaches, our method employs the generator to detect outliers in an online manner. The experimental results show that our method can detect event outliers more accurately than the state-of-the-art approaches.

Ian Gemp, Roma Patel, Yoram Bachrach et al.

Mathematical models of interactions among rational agents have long been studied in game theory. However these interactions are often over a small set of discrete game actions which is very different from how humans communicate in natural language. To bridge this gap, we introduce a framework that allows equilibrium solvers to work over the space of natural language dialogue generated by large language models (LLMs). Specifically, by modelling the players, strategies and payoffs in a "game" of dialogue, we create a binding from natural language interactions to the conventional symbolic logic of game theory. Given this binding, we can ask existing game-theoretic algorithms to provide us with strategic solutions (e.g., what string an LLM should generate to maximize payoff in the face of strategic partners or opponents), giving us predictors of stable, rational conversational strategies. We focus on three domains that require different negotiation strategies: scheduling meetings, trading fruit and debate, and evaluate an LLM's generated language when guided by solvers. We see that LLMs that follow game-theory solvers result in dialogue generations that are less exploitable than the control (no guidance from solvers), and the language generated results in higher rewards, in all negotiation domains. We discuss future implications of this work, and how game-theoretic solvers that can leverage the expressivity of natural language can open up a new avenue of guiding language research.

Heng Wang, Chaoyi Zhang, Jianhui Yu et al.

Automatic 3D neuron reconstruction is critical for analysing the morphology and functionality of neurons in brain circuit activities. However, the performance of existing tracing algorithms is hinged by the low image quality. Recently, a series of deep learning based segmentation methods have been proposed to improve the quality of raw 3D optical image stacks by removing noises and restoring neuronal structures from low-contrast background. Due to the variety of neuron morphology and the lack of large neuron datasets, most of current neuron segmentation models rely on introducing complex and specially-designed submodules to a base architecture with the aim of encoding better feature representations. Though successful, extra burden would be put on computation during inference. Therefore, rather than modifying the base network, we shift our focus to the dataset itself. The encoder-decoder backbone used in most neuron segmentation models attends only intra-volume voxel points to learn structural features of neurons but neglect the shared intrinsic semantic features of voxels belonging to the same category among different volumes, which is also important for expressive representation learning. Hence, to better utilise the scarce dataset, we propose to explicitly exploit such intrinsic features of voxels through a novel voxel-level cross-volume representation learning paradigm on the basis of an encoder-decoder segmentation model. Our method introduces no extra cost during inference. Evaluated on 42 3D neuron images from BigNeuron project, our proposed method is demonstrated to improve the learning ability of the original segmentation model and further enhancing the reconstruction performance.

Tiange Xiang, Chaoyi Zhang, Yang Song et al.

Regularizers help deep neural networks prevent feature co-adaptations. Dropout, as a commonly used regularization technique, stochastically disables neuron activations during network optimization. However, such complete feature disposal can affect the feature representation and network understanding. Toward better descriptions of latent representations, we present DropGraph that learns a regularization function by constructing a stand-alone graph from the backbone features. DropGraph first samples stochastic spatial feature vectors and then incorporates graph reasoning methods to generate feature map distortions. This add-on graph regularizes the network during training and can be completely skipped during inference. We provide intuitions on the linkage between graph reasoning and Dropout with further discussions on how partial graph reasoning method reduces feature correlations. To this end, we extensively study the modeling of graph vertex dependencies and the utilization of the graph for distorting backbone feature maps. DropGraph was validated on 4 tasks with a total of 8 different datasets. The experimental results show that our method outperforms other state-of-the-art regularizers while leaving the base model structure unmodified during inference.

Heng Wang, Yang Song, Chaoyi Zhang et al.

One of the critical steps in improving accurate single neuron reconstruction from three-dimensional (3D) optical microscope images is the neuronal structure segmentation. However, they are always hard to segment due to the lack in quality. Despite a series of attempts to apply convolutional neural networks (CNNs) on this task, noise and disconnected gaps are still challenging to alleviate with the neglect of the non-local features of graph-like tubular neural structures. Hence, we present an end-to-end segmentation network by jointly considering the local appearance and the global geometry traits through graph reasoning and a skeleton-based auxiliary loss. The evaluation results on the Janelia dataset from the BigNeuron project demonstrate that our proposed method exceeds the counterpart algorithms in performance.

Zhaowei Zhu, Xiang Lan, Tingting Zhao et al.

Cardiovascular disease is a major threat to health and one of the primary causes of death globally. The 12-lead ECG is a cheap and commonly accessible tool to identify cardiac abnormalities. Early and accurate diagnosis will allow early treatment and intervention to prevent severe complications of cardiovascular disease. In the PhysioNet/Computing in Cardiology Challenge 2020, our objective is to develop an algorithm that automatically identifies 27 ECG abnormalities from 12-lead ECG recordings.

Carlo Curino, Neha Godwal, Brian Kroth et al.

Developing modern systems software is a complex task that combines business logic programming and Software Performance Engineering (SPE). The later is an experimental and labor-intensive activity focused on optimizing the system for a given hardware, software, and workload (hw/sw/wl) context. Today's SPE is performed during build/release phases by specialized teams, and cursed by: 1) lack of standardized and automated tools, 2) significant repeated work as hw/sw/wl context changes, 3) fragility induced by a "one-size-fit-all" tuning (where improvements on one workload or component may impact others). The net result: despite costly investments, system software is often outside its optimal operating point - anecdotally leaving 30% to 40% of performance on the table. The recent developments in Data Science (DS) hints at an opportunity: combining DS tooling and methodologies with a new developer experience to transform the practice of SPE. In this paper we present: MLOS, an ML-powered infrastructure and methodology to democratize and automate Software Performance Engineering. MLOS enables continuous, instance-level, robust, and trackable systems optimization. MLOS is being developed and employed within Microsoft to optimize SQL Server performance. Early results indicated that component-level optimizations can lead to 20%-90% improvements when custom-tuning for a specific hw/sw/wl, hinting at a significant opportunity. However, several research challenges remain that will require community involvement. To this end, we are in the process of open-sourcing the MLOS core infrastructure, and we are engaging with academic institutions to create an educational program around Software 2.0 and MLOS ideas.

Donghao Zhang, Siqi Liu, Shikha Chaganti et al.

With the injection of contrast material into blood vessels, multi-phase contrasted CT images can enhance the visibility of vessel networks in the human body. Reconstructing the 3D geometric morphology of liver vessels from the contrasted CT images can enable multiple liver preoperative surgical planning applications. Automatic reconstruction of liver vessel morphology remains a challenging problem due to the morphological complexity of liver vessels and the inconsistent vessel intensities among different multi-phase contrasted CT images. On the other side, high integrity is required for the 3D reconstruction to avoid decision making biases. In this paper, we propose a framework for liver vessel morphology reconstruction using both a fully convolutional neural network and a graph attention network. A fully convolutional neural network is first trained to produce the liver vessel centerline heatmap. An over-reconstructed liver vessel graph model is then traced based on the heatmap using an image processing based algorithm. We use a graph attention network to prune the false-positive branches by predicting the presence probability of each segmented branch in the initial reconstruction using the aggregated CNN features. We evaluated the proposed framework on an in-house dataset consisting of 418 multi-phase abdomen CT images with contrast. The proposed graph network pruning improves the overall reconstruction F1 score by 6.4% over the baseline. It also outperformed the other state-of-the-art curvilinear structure reconstruction algorithms.

Siqi Liu, Arnaud Arindra Adiyoso Setio, Florin C. Ghesu et al.

Detecting malignant pulmonary nodules at an early stage can allow medical interventions which may increase the survival rate of lung cancer patients. Using computer vision techniques to detect nodules can improve the sensitivity and the speed of interpreting chest CT for lung cancer screening. Many studies have used CNNs to detect nodule candidates. Though such approaches have been shown to outperform the conventional image processing based methods regarding the detection accuracy, CNNs are also known to be limited to generalize on under-represented samples in the training set and prone to imperceptible noise perturbations. Such limitations can not be easily addressed by scaling up the dataset or the models. In this work, we propose to add adversarial synthetic nodules and adversarial attack samples to the training data to improve the generalization and the robustness of the lung nodule detection systems. To generate hard examples of nodules from a differentiable nodule synthesizer, we use projected gradient descent (PGD) to search the latent code within a bounded neighbourhood that would generate nodules to decrease the detector response. To make the network more robust to unanticipated noise perturbations, we use PGD to search for noise patterns that can trigger the network to give over-confident mistakes. By evaluating on two different benchmark datasets containing consensus annotations from three radiologists, we show that the proposed techniques can improve the detection performance on real CT data. To understand the limitations of both the conventional networks and the proposed augmented networks, we also perform stress-tests on the false positive reduction networks by feeding different types of artificially produced patches. We show that the augmented networks are more robust to both under-represented nodules as well as resistant to noise perturbations.

Paul Muller, Shayegan Omidshafiei, Mark Rowland et al.

This paper investigates a population-based training regime based on game-theoretic principles called Policy-Spaced Response Oracles (PSRO). PSRO is general in the sense that it (1) encompasses well-known algorithms such as fictitious play and double oracle as special cases, and (2) in principle applies to general-sum, many-player games. Despite this, prior studies of PSRO have been focused on two-player zero-sum games, a regime wherein Nash equilibria are tractably computable. In moving from two-player zero-sum games to more general settings, computation of Nash equilibria quickly becomes infeasible. Here, we extend the theoretical underpinnings of PSRO by considering an alternative solution concept, $α$-Rank, which is unique (thus faces no equilibrium selection issues, unlike Nash) and applies readily to general-sum, many-player settings. We establish convergence guarantees in several games classes, and identify links between Nash equilibria and $α$-Rank. We demonstrate the competitive performance of $α$-Rank-based PSRO against an exact Nash solver-based PSRO in 2-player Kuhn and Leduc Poker. We then go beyond the reach of prior PSRO applications by considering 3- to 5-player poker games, yielding instances where $α$-Rank achieves faster convergence than approximate Nash solvers, thus establishing it as a favorable general games solver. We also carry out an initial empirical validation in MuJoCo soccer, illustrating the feasibility of the proposed approach in another complex domain.

H. Francis Song, Abbas Abdolmaleki, Jost Tobias Springenberg et al.

Some of the most successful applications of deep reinforcement learning to challenging domains in discrete and continuous control have used policy gradient methods in the on-policy setting. However, policy gradients can suffer from large variance that may limit performance, and in practice require carefully tuned entropy regularization to prevent policy collapse. As an alternative to policy gradient algorithms, we introduce V-MPO, an on-policy adaptation of Maximum a Posteriori Policy Optimization (MPO) that performs policy iteration based on a learned state-value function. We show that V-MPO surpasses previously reported scores for both the Atari-57 and DMLab-30 benchmark suites in the multi-task setting, and does so reliably without importance weighting, entropy regularization, or population-based tuning of hyperparameters. On individual DMLab and Atari levels, the proposed algorithm can achieve scores that are substantially higher than has previously been reported. V-MPO is also applicable to problems with high-dimensional, continuous action spaces, which we demonstrate in the context of learning to control simulated humanoids with 22 degrees of freedom from full state observations and 56 degrees of freedom from pixel observations, as well as example OpenAI Gym tasks where V-MPO achieves substantially higher asymptotic scores than previously reported.

Jie Yang, Siqi Liu, Sasa Grbic et al.

Though large-scale datasets are essential for training deep learning systems, it is expensive to scale up the collection of medical imaging datasets. Synthesizing the objects of interests, such as lung nodules, in medical images based on the distribution of annotated datasets can be helpful for improving the supervised learning tasks, especially when the datasets are limited by size and class balance. In this paper, we propose the class-aware adversarial synthesis framework to synthesize lung nodules in CT images. The framework is built with a coarse-to-fine patch in-painter (generator) and two class-aware discriminators. By conditioning on the random latent variables and the target nodule labels, the trained networks are able to generate diverse nodules given the same context. By evaluating on the public LIDC-IDRI dataset, we demonstrate an example application of the proposed framework for improving the accuracy of the lung nodule malignancy estimation as a binary classification problem, which is important in the lung screening scenario. We show that combining the real image patches and the synthetic lung nodules in the training set can improve the mean AUC classification score across different network architectures by 2%.

Siqi Liu, Eli Gibson, Sasa Grbic et al.

The performance of medical image analysis systems is constrained by the quantity of high-quality image annotations. Such systems require data to be annotated by experts with years of training, especially when diagnostic decisions are involved. Such datasets are thus hard to scale up. In this context, it is hard for supervised learning systems to generalize to the cases that are rare in the training set but would be present in real-world clinical practices. We believe that the synthetic image samples generated by a system trained on the real data can be useful for improving the supervised learning tasks in the medical image analysis applications. Allowing the image synthesis to be manipulable could help synthetic images provide complementary information to the training data rather than simply duplicating the real-data manifold. In this paper, we propose a framework for synthesizing 3D objects, such as pulmonary nodules, in 3D medical images with manipulable properties. The manipulation is enabled by decomposing of the object of interests into its segmentation mask and a 1D vector containing the residual information. The synthetic object is refined and blended into the image context with two adversarial discriminators. We evaluate the proposed framework on lung nodules in 3D chest CT images and show that the proposed framework could generate realistic nodules with manipulable shapes, textures and locations, etc. By sampling from both the synthetic nodules and the real nodules from 2800 3D CT volumes during the classifier training, we show the synthetic patches could improve the overall nodule detection performance by average 8.44% competition performance metric (CPM) score.

Siqi Liu, Daguang Xu, S. Kevin Zhou et al.

While deep convolutional neural networks (CNN) have been successfully applied for 2D image analysis, it is still challenging to apply them to 3D anisotropic volumes, especially when the within-slice resolution is much higher than the between-slice resolution and when the amount of 3D volumes is relatively small. On one hand, direct learning of CNN with 3D convolution kernels suffers from the lack of data and likely ends up with poor generalization; insufficient GPU memory limits the model size or representational power. On the other hand, applying 2D CNN with generalizable features to 2D slices ignores between-slice information. Coupling 2D network with LSTM to further handle the between-slice information is not optimal due to the difficulty in LSTM learning. To overcome the above challenges, we propose a 3D Anisotropic Hybrid Network (AH-Net) that transfers convolutional features learned from 2D images to 3D anisotropic volumes. Such a transfer inherits the desired strong generalization capability for within-slice information while naturally exploiting between-slice information for more effective modelling. The focal loss is further utilized for more effective end-to-end learning. We experiment with the proposed 3D AH-Net on two different medical image analysis tasks, namely lesion detection from a Digital Breast Tomosynthesis volume, and liver and liver tumor segmentation from a Computed Tomography volume and obtain the state-of-the-art results.

Siqi Liu, Zhenhai Zhu, Ning Ye et al.

Current image captioning methods are usually trained via (penalized) maximum likelihood estimation. However, the log-likelihood score of a caption does not correlate well with human assessments of quality. Standard syntactic evaluation metrics, such as BLEU, METEOR and ROUGE, are also not well correlated. The newer SPICE and CIDEr metrics are better correlated, but have traditionally been hard to optimize for. In this paper, we show how to use a policy gradient (PG) method to directly optimize a linear combination of SPICE and CIDEr (a combination we call SPIDEr): the SPICE score ensures our captions are semantically faithful to the image, while CIDEr score ensures our captions are syntactically fluent. The PG method we propose improves on the prior MIXER approach, by using Monte Carlo rollouts instead of mixing MLE training with PG. We show empirically that our algorithm leads to easier optimization and improved results compared to MIXER. Finally, we show that using our PG method we can optimize any of the metrics, including the proposed SPIDEr metric which results in image captions that are strongly preferred by human raters compared to captions generated by the same model but trained to optimize MLE or the COCO metrics.

Siqi Liu, Sidong Liu, Sonia Pujol et al.

We present a longitudinal MR simulation framework which simulates the future neurodegenerative progression by outputting the predicted follow-up MR image and the voxel-based morphometry (VBM) map. This framework expects the patients to have at least 2 historical MR images available. The longitudinal and cross-sectional VBM maps are extracted to measure the affinity between the target subject and the template subjects collected for simulation. Then the follow-up simulation is performed by resampling the latest available target MR image with a weighted sum of non-linear transformations derived from the best-matched templates. The leave-one-out strategy was used to compare different simulation methods. Compared to the state-of-the-art voxel-based method, our proposed morphometry-based simulation achieves better accuracy in most cases.