论文检索

Zhuo Chen, Haimei Zhao, Bo Yuan, Xiu Li

Multi-camera depth estimation has recently garnered significant attention due to its substantial practical implications in the realm of autonomous driving. In this paper, we delve into the task of self-supervised multi-camera depth estimation and propose an innovative framework, STViT, featuring several noteworthy enhancements: 1) we propose a Spatial-Temporal Transformer to comprehensively exploit both local connectivity and the global context of image features, meanwhile learning enriched spatial-temporal cross-view correlations to recover 3D geometry. 2) to alleviate the severe effect of adverse conditions, e.g., rainy weather and nighttime driving, we introduce a GAN-based Adversarial Geometry Regularization Module (AGR) to further constrain the depth estimation with unpaired normal-condition depth maps and prevent the model from being incorrectly trained. Experiments on challenging autonomous driving datasets Nuscenes and DDAD show that our method achieves state-of-the-art performance.

Zhuo Chen, Haimei Zhao, Chaoyue Wang, Bo Yuan, Xiu Li

3D-aware GANs successfully solve the problem of 3D-consistency generation and furthermore provide a 3D shape of the generated object. However, the application of the volume renderer disturbs the disentanglement of the latent space, which makes it difficult to manipulate 3D-aware GANs and lowers the image quality of style-based generators. In this work, we devise a dual-mapping framework to make the generated images of pretrained 2D StyleGAN consistent in 3D space. We utilize a tri-plane representation to estimate the 3D shape of the generated object and two mapping networks to bridge the latent space of StyleGAN and the 3D tri-plane space. Our method does not alter the parameters of the pretrained generator, which means the interpretability of latent space is preserved for various image manipulations. Experiments show that our method lifts the 3D awareness of pretrained 2D StyleGAN to 3D-aware GANs and outperforms the 3D-aware GANs in controllability and image quality.

Xiaojian Chen, Chuyue Liao, Yanhui Gu, Yafei Li, Jinlan Wang, Yi Chen, Masaru Kitsuregawa

Matching molecular analogues is a computational chemistry and bioinformatics research issue which is used to identify molecules that are structurally or functionally similar to a target molecule. Recent studies on matching analogous molecules have predominantly concentrated on enhancing effectiveness, often sidelining computational efficiency, particularly in contexts of low computational resources. This oversight poses challenges in many real applications (e.g., drug discovery, catalyst generation and so forth). To tackle this issue, we propose a general strategy named MapLE, aiming to promptly match analogous molecules with low computational resources by multi-metrics evaluation. Experimental evaluation conducted on a public biomolecular dataset validates the excellent and efficient performance of the proposed strategy.

Tianyi Chen, Feiqi Cao, Yihao Ding, Caren Han

With the introduction of large language models, chatbots are becoming more conversational to communicate effectively and capable of handling increasingly complex tasks. To make a chatbot more relatable and engaging, we propose a new language model idea that maps the human-like personality. In this paper, we propose a systematic Personality-Enhanced Language Model (PELM) approach by using a joint learning mechanism of personality classification and language generation tasks. The proposed PELM leverages a dataset of defined personality typology, Myers-Briggs Type Indicator, and produces a Personality-Enhanced Language Model by using a joint learning and cross-teaching structure consisting of a classification and language modelling to incorporate personalities via both distinctive types and textual information. The results show that PELM can generate better personality-based outputs than baseline models.

Bin Chen, Kai Yang, Wenxin Tai, Zhangtao Cheng, Leyuan Liu, Ting Zhong, Fan Zhou

Temporal knowledge graph reasoning is an essential task that holds immense value in diverse real-world applications. Existing studies mainly focus on leveraging structural and sequential dependencies, excelling in tasks like entity and link prediction. However, they confront a notable interpretability gap in their predictions, a pivotal facet for comprehending model behavior. In this study, we propose an innovative method, LSGAT, which not only exhibits remarkable precision in entity predictions but also enhances interpretability by identifying pivotal historical events influencing event predictions. LSGAT enables concise explanations for prediction outcomes, offering valuable insights into the otherwise enigmatic "black box" reasoning process. Through an exploration of the implications of the most influential events, it facilitates a deeper understanding of the underlying mechanisms governing predictions.

Angela Chen, Nicholas Gisolfi, Artur Dubrawski

Ensuring a machine learning model’s trustworthiness is crucial to prevent potential harm. One way to foster trust is through the formal verification of the model’s adherence to essential design requirements. However, this approach relies on well-defined, application-domain-centric criteria with which to test the model, and such specifications may be cumbersome to collect in practice. We propose a data-driven approach for creating specifications to evaluate a trained model effectively. Implementing this framework allows us to prove that the model will exhibit safe behavior while minimizing the false-positive prediction rate. This strategy enhances predictive accuracy and safety, providing deeper insight into the model’s strengths and weaknesses, and promotes trust through a systematic approach.

Yifu Cai, Arvind Srinivasan, Mononito Goswami, Arjun Choudhry, Artur Dubrawski

Time-series and text data are prevalent in healthcare and frequently co-exist, yet they are typically modeled in isolation. Even studies that jointly model time-series and text, do so by converting time-series to images or graphs. We hypothesize that explicitly modeling time-series jointly with text can improve tasks such as summarization and question answering for time-series data, which have received little attention so far. To address this gap, we introduce JoLT to jointly learn desired representations from pre-trained time-series and text models. JoLT utilizes a Querying Transformer (Q-Former) to align the time-series and text representations. Our experiments on a large real-world electrocardiography dataset for medical time-series summarization show that JoLT outperforms state-of-the-art image captioning approaches.

Rickard Brännvall

To enhance the computational efficiency of quantized Transformers, we replace the dot-product and Softmax-based attention with an alternative mechanism involving addition and ReLU activation only. This side-steps the expansion to double precision often required by matrix multiplication and avoids costly Softmax evaluations but maintains much of the core functionality of conventional dot-product attention. It can enable more efficient execution and support larger quantized Transformer models on resource-constrained hardware or alternative arithmetic systems like homomorphic encryption. Training experiments on four common benchmark tasks show test set prediction scores comparable to those of conventional Transformers with dot-product attention. Our scaling experiments also suggest significant computational savings, both in plaintext and under encryption. In particular, we believe that the ReLU and addition-based attention mechanism introduced in this paper may enable privacy-preserving AI applications operating under homomorphic encryption by avoiding the costly multiplication of encrypted variables.

Tuhin Kumar Biswas, Avisek Gupta, Narayan Changder, Redha Taguelmimt, Samir Aknine, Samiran Chattopadhyay, Animesh Dutta

Simultaneous Coalition Structure Generation and Assignment (SCSGA) is an important research problem in multi-agent systems. Given n agents and m tasks, the aim of SCSGA is to form m disjoint coalitions of n agents such that between the coalitions and tasks there is a one-to-one mapping, which ensures each coalition is capable of accomplishing the assigned task. SCSGA with Multi-dimensional Features (SCSGA-MF) extends the problem by introducing a d-dimensional vector for each agent and task. We propose a heuristic algorithm called Multiple Distance Metric (MDM) approach to solve SCSGA-MF. Experimental results confirm that MDM produces near optimal solutions, while being feasible for large-scale inputs within a reasonable time frame.

Anthony Bazhenov, Pahan Dewasurendra, Giri Krishnan, Jean Erik Delanois

The performance of artificial neural networks (ANNs) degrades when training data are limited or imbalanced. In contrast, the human brain can learn quickly from just a few examples. Here, we investigated the role of sleep in improving the performance of ANNs trained with limited data on the MNIST and Fashion MNIST datasets. Sleep was implemented as an unsupervised phase with local Hebbian type learning rules. We found a significant boost in accuracy after the sleep phase for models trained with limited data in the range of 0.5-10% of total MNIST or Fashion MNIST datasets. When more than 10% of the total data was used, sleep alone had a slight negative impact on performance, but this was remedied by fine-tuning on the original data. This study sheds light on a potential synaptic weight dynamics strategy employed by the brain during sleep to enhance memory performance when training data are limited or imbalanced.

Amine Barrak

Distributed ML addresses challenges from increasing data and model complexities. Peer to peer (P2P) networks in distributed ML offer scalability and fault tolerance. However, they also encounter challenges related to resource consumption, and communication overhead as the number of participating peers grows. This research introduces a novel architecture that combines serverless computing with P2P networks for distributed training. Serverless computing enhances this model with parallel processing and cost effective scalability, suitable for resource-intensive tasks. Preliminary results show that peers can offload expensive computational tasks to serverless platforms. However, their inherent statelessness necessitates strong communication methods, suggesting a pivotal role for databases. To this end, we have enhanced an in memory database to support ML training tasks.

Alaleh Ahmadianshalchi, Syrine Belakaria, Janardhan Rao Doppa

We consider the problem of constrained multi-objective optimization over black-box objectives, with user-defined preferences, with a largely infeasible input space. Our goal is to approximate the optimal Pareto set from the small fraction of feasible inputs. The main challenges include huge design space, multiple objectives, numerous constraints, and rare feasible inputs identified only through expensive experiments. We present PAC-MOO, a novel preference-aware multi-objective Bayesian optimization algorithm to solve this problem. It leverages surrogate models for objectives and constraints to intelligently select the sequence of inputs for evaluation to achieve the target goal.

Adin Aberbach, Mayank Kejriwal, Ke Shen

Entity Resolution (ER) is the problem of algorithmically matching records, mentions, or entries that refer to the same underlying real-world entity. Traditionally, the problem assumes (at most) two datasets, between which records need to be matched. There is considerably less research in ER when k > 2 datasets are involved. The evaluation of such multipartite ER (M-ER) is especially complex, since the usual ER metrics assume (whether implicitly or explicitly) k < 3. This paper takes the first step towards motivating a k-tuple approach for evaluating M-ER. Using standard algorithms and k-tuple versions of metrics like precision and recall, our preliminary results suggest a significant difference compared to aggregated pairwise evaluation, which would first decompose the M-ER problem into independent bipartite problems and then aggregate their metrics. Hence, M-ER may be more challenging and warrant more novel approaches than current decomposition-based pairwise approaches would suggest.

Yuan Yang

Despite current AI’s human-like behavior, super efficiency, and unbelievable ability to handle complex games, we still complain that it shows no sign of creativity, originality, or novelty outside its training set, and that it fails to develop new insights into old experience or establish understanding of new experience. In short, it generates content from its training set, but does not invent content. A fundamental reason for this is that current AI is incapable of abstraction and reasoning in an abstract, generalizable, and systematic way. Think, for instance, of what AI systems we can build if we have a base system that can answer this simple question—when two things are the same. Instead of studying these high-level questions, I put my thesis in the context of visual abstract reasoning (VAR), a task widely used in human intelligence tests. A classical example of this task is Raven’s Progressive Matrices (RPM, see Figure 1), a family of intelligence tests that was designed to measure eductive ability, i.e., the ability to make meaning out of confusion and generate high-level, usually nonverbal, schemata which make it easy to handle complexity. A similar concept to eductive ability is fluid intelligence, or the ability to discriminate and perceive complex relationships when no recourse to answers is stored in memory. Whether eductive ability or fluid intelligence, RPM points to the qualities that have been lacking in AI. To explore these qualities in AI, I propose the following research questions.

Luisa Werner

The goal of this thesis is to address knowledge graph completion tasks using neuro-symbolic methods. Neuro-symbolic methods allow the joint utilization of symbolic information defined as meta-rules in ontologies and knowledge graph embedding methods that represent entities and relations of the graph in a low-dimensional vector space. This approach has the potential to improve the resolution of knowledge graph completion tasks in terms of reliability, interpretability, data-efficiency and robustness.

Pulkit Verma

The vast diversity of internal designs of taskable black-box AI systems and their nuanced zones of safe functionality make it difficult for a layperson to use them without unintended side effects. My dissertation focuses on developing paradigms that enable a user to assess and understand the limits of an AI system's safe operability. We develop a personalized AI assessment module that lets an AI system execute instruction sequences in simulators and answer queries about these executions. Our results show that such a primitive query-response interface is sufficient to efficiently derive a user-interpretable model of a system's capabilities.

Fiona Anting Tan

Causality expresses the relation between two arguments, one of which represents the cause and the other the effect (or consequence). Causal text mining refers to the extraction and usage of causal information from text. Given an input sequence, we are interested to know if and where causal information occurs. My research is focused on the end-to-end challenges of causal text mining. This involves extracting, representing, and applying causal knowledge from unstructured text. The corresponding research questions are: (1) How to extract causal information from unstructured text effectively? (2) How to represent extracted causal relationships in a graph that is interpretable and useful for some application? (3) How can we capitalize on extracted causal knowledge for downstream tasks? What tasks or fields will benefit from such knowledge? In this paper, I outline past and on-going works, and highlight future research challenges.

Aaquib Tabrez

Policy explanation, a process for describing the behavior of an autonomous system, plays a crucial role in effectively conveying an agent's decision-making rationale to human collaborators and is essential for safe real-world deployments. It becomes even more critical in effective human-robot teaming, where good communication allows teams to adapt and improvise successfully during uncertain situations by enabling value alignment within the teams. This thesis proposal focuses on improving human-machine teaming by developing novel human-centered explainable AI (xAI) techniques that empower autonomous agents to communicate their capabilities and limitations via multiple modalities, teach and influence human teammates' behavior as decision-support systems, and effectively build and manage trust in HRI systems.

Nisha Simon

Humans have been using stories to entertain, educate, and persuade audiences for centuries. The advent of modern AI tools in the form of Large Language Models (LLMs) such as chatGPT continues to fulfill this purpose. However while recent work has shown that LLMs can successfully be used for narrative generation, they lack coherence and can be prone to repetition and stilted language. Automated Planning can therefore be combined with Natural Language text generation to create narratives (stories) that are logical, coherent, and believable. A planning model provides scaffolding to an LLM so that the LLM's language generation is context-dependent, in order to allow users to create more coherent, logical, and believable stories in a variety of domains.

Ke Shen

The advent of powerful transformer-based discriminative language models and, more recently, generative GPT-family models, has led to notable advancements in natural language processing (NLP), particularly in commonsense reasoning tasks. One such task is commonsense reasoning, where performance is usually evaluated through multiple-choice question-answering benchmarks. Till date, many such benchmarks have been proposed and `leaderboards' tracking state-of-the-art performance on those benchmarks suggest that transformer-based models are approaching human-like performance. However, due to documented problems such as hallucination and bias, the research focus is shifting from merely quantifying accuracy on the task to an in-depth, context-sensitive probing of LLMs' generalization and robustness. To gain deeper insight into diagnosing these models' performance in commonsense reasoning scenarios, this thesis addresses three main studies: the generalization ability of transformer-based language models on commonsense reasoning, the trend in confidence distribution of these language models confronted with ambiguous inference tasks, and a proposed risk-centric evaluation framework for both discriminative and generative language models.