As the pre-training objectives (e.g., next token prediction) of language models (LMs) are inherently not aligned with task scores, optimizing LMs to achieve higher downstream task scores is essential. One of the promising approaches is to fine-tune LMs through reinforcement learning (RL). However, conventional RL methods based on PPO and a penalty of KL divergence are vulnerable to text degeneration where LMs do not generate natural texts anymore after RL fine-tuning. To address this problem, we provide Degeneration-free Policy Optimization (DfPO) that can fine-tune LMs to generate texts that achieve improved downstream task scores, while preserving the ability to generate natural texts. To achieve this, we introduce KL-masking which masks out the actions that potentially cause deviation from the reference policy when its likelihood is increased or decreased. Then, we devise truncated advantage functions for separately performing likelihood maximization and minimization to improve the task performance. In the experiments, we provide the results of DfPO and baseline algorithms on various generative NLP tasks including text continuation, text detoxification, and commonsense generation. Our experiments demonstrate that DfPO successfully improves the downstream task scores while preserving the ability to generate natural texts, without requiring additional hyperparameter search.

We consider offline safe imitation learning (IL), where the agent aims to learn the safe policy that mimics preferred behavior while avoiding non-preferred behavior from non-preferred demonstrations and unlabeled demonstrations. This problem setting corresponds to various real-world scenarios, where satisfying safety constraints is more important than maximizing the expected return. However, it is very challenging to learn the policy to avoid constraint-violating (i.e. non-preferred) behavior, as opposed to standard imitation learning which learns the policy to mimic given demonstrations. In this paper, we present a hyperparameter-free offline safe IL algorithm, SafeDICE, that learns safe policy by leveraging the non-preferred demonstrations in the space of stationary distributions. Our algorithm directly estimates the stationary distribution corrections of the policy that imitate the demonstrations excluding the non-preferred behavior. In the experiments, we demonstrate that our algorithm learns a more safe policy that satisfies the cost constraint without degrading the reward performance, compared to baseline algorithms.

The recent advances in representation learning inspire us to take on the challenging problem of unsupervised image classification tasks in a principled way. We propose ContraCluster, an unsupervised image classification method that combines clustering with the power of contrastive self-supervised learning. ContraCluster consists of three stages: (1) contrastive self-supervised pre-training (CPT), (2) contrastive prototype sampling (CPS), and (3) prototype-based semi-supervised fine-tuning (PB-SFT). CPS can select highly accurate, categorically prototypical images in an embedding space learned by contrastive learning. We use sampled prototypes as noisy labeled data to perform semi-supervised fine-tuning (PB-SFT), leveraging small prototypes and large unlabeled data to further enhance the accuracy. We demonstrate empirically that ContraCluster achieves new state-of-the-art results for standard benchmark datasets including CIFAR-10, STL-10, and ImageNet-10. For example, ContraCluster achieves about 90.8% accuracy for CIFAR-10, which outperforms DAC (52.2%), IIC (61.7%), and SCAN (87.6%) by a large margin. Without any labels, ContraCluster can achieve a 90.8% accuracy that is comparable to 95.8% by the best supervised counterpart.

Despite surprising performance on zero-shot transfer, pre-training a large-scale multimodal model is often prohibitive as it requires a huge amount of data and computing resources. In this paper, we propose a method (BeamCLIP) that can effectively transfer the representations of a large pre-trained multimodal model (CLIP-ViT) into a small target model (e.g., ResNet-18). For unsupervised transfer, we introduce cross-modal similarity matching (CSM) that enables a student model to learn the representations of a teacher model by matching the relative similarity distribution across text prompt embeddings. To better encode the text prompts, we design context-based prompt augmentation (CPA) that can alleviate the lexical ambiguity of input text prompts. Our experiments show that unsupervised representation transfer of a pre-trained vision-language model enables a small ResNet-18 to achieve a better ImageNet-1K top-1 linear probe accuracy (66.2%) than vision-only self-supervised learning (SSL) methods (e.g., SimCLR: 51.8%, SwAV: 63.7%), while closing the gap with supervised learning (69.8%).

Active Learning for discriminative models has largely been studied with the focus on individual samples, with less emphasis on how classes are distributed or which classes are hard to deal with. In this work, we show that this is harmful. We propose a method based on the Bayes’ rule, that can naturally incorporate class imbalance into the Active Learning framework. We derive that three terms should be considered together when estimating the probability of a classifier making a mistake for a given sample; i) probability of mislabelling a class, ii) likelihood of the data given a predicted class, and iii) the prior probability on the abundance of a predicted class. Implementing these terms requires a generative model and an intractable likelihood estimation. Therefore, we train a Variational Auto Encoder (VAE) for this purpose. To further tie the VAE with the classifier and facilitate VAE training, we use the classifiers’ deep feature representations as input to the VAE. By considering all three probabilities, among them, especially the data imbalance, we can substantially improve the potential of existing methods under limited data budget. We show that our method can be applied to classification tasks on multiple different datasets – including one that is a real-world dataset with heavy data imbalance – significantly outperforming the state of the art.

We propose DefogGAN, a generative approach to the problem of inferring state information hidden in the fog of war for real-time strategy (RTS) games. Given a partially observed state, DefogGAN generates defogged images of a game as predictive information. Such information can lead to create a strategic agent for the game. DefogGAN is a conditional GAN variant featuring pyramidal reconstruction loss to optimize on multiple feature resolution scales. We have validated DefogGAN empirically using a large dataset of professional StarCraft replays. Our results indicate that DefogGAN can predict the enemy buildings and combat units as accurately as professional players do and achieves a superior performance among state-of-the-art defoggers.

In neural combinatorial optimization (CO), reinforcement learning (RL) can turn a deep neural net into a fast, powerful heuristic solver of NP-hard problems. This approach has a great potential in practical applications because it allows near-optimal solutions to be found without expert guides armed with substantial domain knowledge. We introduce Policy Optimization with Multiple Optima (POMO), an end-to-end approach for building such a heuristic solver. POMO is applicable to a wide range of CO problems. It is designed to exploit the symmetries in the representation of a CO solution. POMO uses a modified REINFORCE algorithm that forces diverse rollouts towards all optimal solutions. Empirically, the low-variance baseline of POMO makes RL training fast and stable, and it is more resistant to local minima compared to previous approaches. We also introduce a new augmentation-based inference method, which accompanies POMO nicely. We demonstrate the effectiveness of POMO by solving three popular NP-hard problems, namely, traveling salesman (TSP), capacitated vehicle routing (CVRP), and 0-1 knapsack (KP). For all three, our solver based on POMO shows a significant improvement in performance over all recent learned heuristics. In particular, we achieve the optimality gap of 0.14% with TSP100 while reducing inference time by more than an order of magnitude.

In this paper, we introduce Activity Travel Pattern (ATP) monitoring in a large-scale city environment. ATP represents where city residents and vehicles stay and how they travel around in a complex megacity. Monitoring ATP will incubate new types of value-added services such as predictive mobile advertisement, demand forecasting for urban stores, and adaptive transportation scheduling. To enable ATP monitoring, we develop ActraMon, a high-performanceATP monitoring framework. As a first step, ActraMon provides a simple but effective computational model of ATP and a declarative query language facilitating effective specification of various ATP monitoring queries. More important, ActraMon employs the shared staging architecture and highly efficient processing techniques, which address the scalability challenges caused by massive location updates, a number of ATP monitoring queries and processing complexity of ATP monitoring. Finally, we demonstrate the extensive performance study of ActraMon using realistic city-wide ATP workloads.

Border Monitoring Query (BMQ) has different query semantic from conventional continuous range query. It monitors the values of data streams and reports them only when data streams cross the borders of its range. In this paper, we first emphasize the importance and usefulness of BMQ through attractive service scenarios. Then, we propose BMQ-Index, which is specialized to BMQ evaluation. It efficiently processes a large number of BMQs in a shared and incremental manner. For shared processing, BMQ-Index adopts a query indexing approach, thereby achieving a high level of scalability. For incremental processing, BMQ-Index employs an incremental access method. Thus, successive BMQ evaluations are significantly accelerated. We present an index structure and a search algorithm to support onedimensional as well as multi-dimensional BMQ. Lastly, we demonstrate the performance benefits of BMQ-Index through analysis and experiments.