Modern research computing cyberinfrastructure, such as ACCESS-CI and NAIRR Pilot, offers GPU resources across geographically distributed clusters to accommodate the increasing needs of scientific deep learning (DL) workloads. Even for high-performance computing (HPC) experts, configuring environments and managing DL workloads across supercomputers remain significant barriers. To address these obstacles, we present Diamond, an open-source platform to simplify and streamline the DL lifecycle on HPC. Diamond provides an intuitive graphical interface that abstracts system-level complexity, enabling users to develop, debug, and deploy DL models with minimal overhead. We identify several challenges in building such a platform, including portability, security, and usability, and propose effective architectural solutions to each. Notably, Diamond enables users to share and reuse DL workload environments across systems and collaborators, reducing redundant setup efforts. Experimental results demonstrate that Diamond reduces the time to first successful deployment by an average of 68%, compared to manual configuration with command lines. The Diamond service is available at https://diamondhpc.ai.

Scaling laws are powerful tools to predict the performance of large language models. However, current scaling laws fall short of accounting for inference costs. In this work, we first show that model architecture affects inference latency, where models of the same size can have up to $3.5\times$ difference in latency. To tackle this challenge, we modify the Chinchilla scaling laws to co-optimize the model parameter count, the number of training tokens, and the model architecture. Due to the reason that models of similar training loss exhibit gaps in downstream evaluation, we also propose a novel method to train inference-efficient models based on the revised scaling laws. We perform extensive empirical studies to fit and evaluate our inference-aware scaling laws. We vary model parameters from 80M to 1B, training tokens from 1.6B to 30B, and model shapes, training 63 models. Guided by our inference-efficient scaling law and model selection method, we release the Morph-1B model, which improves inference latency by $1.8\times$ while maintaining accuracy on downstream tasks compared to open-source models, pushing the Pareto frontier of accuracy-latency tradeoff. Notably, our experiments reveal that wider and shallower models can yield efficiency gains while preserving accuracy.

We study the certifiable robustness of ML classifiers on dirty datasets that could contain missing values. A test point is certifiably robust for an ML classifier if the classifier returns the same prediction for that test point, regardless of which cleaned version (among exponentially many) of the dirty dataset the classifier is trained on. In this paper, we show theoretically that for Naive Bayes Classifiers (NBC) over dirty datasets with missing values: (i) there exists an efficient polynomial time algorithm to decide whether multiple input test points are all certifiably robust over a dirty dataset; and (ii) the data poisoning attack, which aims to make all input test points certifiably non-robust by inserting missing cells to the clean dataset, is in polynomial time for single test points but NP-complete for multiple test points. Extensive experiments demonstrate that our algorithms are efficient and outperform existing baselines.

Community search is a problem that seeks cohesive and connected subgraphs in a graph that satisfy certain topology constraints, e.g., degree constraints. The majority of existing works focus exclusively on the topology and ignore the nodes' influence in the communities. To tackle this deficiency, influential community search is further proposed to include the node's influence. Each node has a weight, namely influence value, in the influential community search problem to represent its network influence. The influence value of a community is produced by an aggregated function, e.g., max, min, avg, and sum, over the influence values of the nodes in the same community. The objective of the influential community search problem is to locate the top-r communities with the highest influence values while satisfying the topology constraints. Existing studies on influential community search have several limitations: (i) they focus exclusively on simple aggregation functions such as min, which may fall short of certain requirements in many real-world scenarios, and (ii) they impose no limitation on the size of the community, whereas most real-world scenarios do. This motivates us to conduct a new study to fill this gap. We consider the problem of identifying the top-r influential communities with/without size constraints while using more complicated aggregation functions such as sum or avg. We give a theoretical analysis demonstrating the hardness of the problems and propose efficient and effective heuristic solutions for our top-r influential community search problems. Extensive experiments on real large graphs demonstrate that our proposed solution is significantly more efficient than baseline solutions.

Text classification is a fundamental task in content analysis. Nowadays, deep learning has demonstrated promising performance in text classification compared with shallow models. However, almost all the existing models do not take advantage of the wisdom of human beings to help text classification. Human beings are more intelligent and capable than machine learning models in terms of understanding and capturing the implicit semantic information from text. In this article, we try to take guidance from human beings to classify text. We propose Crowd-powered learning for Text Classification (CrowdTC for short). We design and post the questions on a crowdsourcing platform to extract keywords in text. Sampling and clustering techniques are utilized to reduce the cost of crowdsourcing. Also, we present an attention-based neural network and a hybrid neural network to incorporate the extracted keywords as human guidance into deep neural networks. Extensive experiments on public datasets confirm that CrowdTC improves the text classification accuracy of neural networks by using the crowd-powered keyword guidance.

Exploratory Data Analysis (EDA) is a crucial step in any data science project. However, existing Python libraries fall short in supporting data scientists to complete common EDA tasks for statistical modeling. Their API design is either too low level, which is optimized for plotting rather than EDA, or too high level, which is hard to specify more fine-grained EDA tasks. In response, we propose DataPrep.EDA, a novel task-centric EDA system in Python. DataPrep.EDA allows data scientists to declaratively specify a wide range of EDA tasks in different granularity with a single function call. We identify a number of challenges to implement DataPrep.EDA, and propose effective solutions to improve the scalability, usability, customizability of the system. In particular, we discuss some lessons learned from using Dask to build the data processing pipelines for EDA tasks and describe our approaches to accelerate the pipelines. We conduct extensive experiments to compare DataPrep.EDA with Pandas-profiling, the state-of-the-art EDA system in Python. The experiments show that DataPrep.EDA significantly outperforms Pandas-profiling in terms of both speed and user experience. DataPrep.EDA is open-sourced as an EDA component of DataPrep: https://github.com/sfu-db/dataprep.

Given a social network G , a cost associated with each node, and a budget B , the budgeted influence maximization (BIM) problem aims to find a set S of nodes, denoted as the seed set, that maximizes the expected number of influenced users under the constraint that the total cost of the users in S is no larger than B. The current state-of-the-art practical solution for BIM problem provides a (1-1/ e /2 --- ε)-approximate (≈ 0.316 --- ε) result and is still inefficient on large networks. We first show that we can improve the approximation guarantee to 1 --- 1/ e β --- ε where 1 --- 1/ e β = (1 --- β) (1 --- 1/ e ), achieving a better approximation guarantee (≈ 0.355 --- ε). Next, we apply the reverse sampling based technique, a popular technique for classic influence maximization, to our studied BIM problem. However, it is non-trivial to design efficient solutions for large scale networks even the reverse sampling based technique is applied. On one hand, it is unclear how to derive tight bounds for the nodes selected by the greedy algorithm under the budgeted scenario, where each time it selects the seed node with the highest benefit-cost ratio. With tighter bounds, the algorithm can terminate as soon as the approximation ratio is satisfied, thus saving the running cost. On the other hand, the number of nodes selected under BIM problem may be quite large since it may greedily select many nodes with large benefit-cost ratio but with low costs. The time complexity of existing influence maximization algorithms heavily depends on the size of the seed set. To tackle such challenging issues, we first present new bound estimation techniques for the BIM problem. Next, we present new node selection strategies to alleviate the dependency to the size of the seed set. Extensive experiments show that our proposed solution is far more efficient than alternatives.

Graph similarity search is among the most important graph-based applications, e.g. finding the chemical compounds that are most similar to a query compound. Graph similarity/distance computation, such as Graph Edit Distance (GED) and Maximum Common Subgraph (MCS), is the core operation of graph similarity search and many other applications, but very costly to compute in practice. Inspired by the recent success of neural network approaches to several graph applications, such as node or graph classification, we propose a novel neural network based approach to address this classic yet challenging graph problem, aiming to alleviate the computational burden while preserving a good performance. The proposed approach, called SimGNN, combines two strategies. First, we design a learnable embedding function that maps every graph into an embedding vector, which provides a global summary of a graph. A novel attention mechanism is proposed to emphasize the important nodes with respect to a specific similarity metric. Second, we design a pairwise node comparison method to supplement the graph-level embeddings with fine-grained node-level information. Our model achieves better generalization on unseen graphs, and in the worst case runs in quadratic time with respect to the number of nodes in two graphs. Taking GED computation as an example, experimental results on three real graph datasets demonstrate the effectiveness and efficiency of our approach. Specifically, our model achieves smaller error rate and great time reduction compared against a series of baselines, including several approximation algorithms on GED computation, and many existing graph neural network based models. Our study suggests SimGNN provides a new direction for future research on graph similarity computation and graph similarity search.