Deep learning has improved state-of-the-art results in many important fields, and has been the subject of much research in recent years, leading to the development of several systems for facilitating deep learning. Current systems, however, mainly focus on model building and training phases, while the issues of data management, model sharing, and lifecycle management are largely ignored. Deep learning modeling lifecycle generates a rich set of data artifacts, e.g., learned parameters and training logs, and it comprises of several frequently conducted tasks, e.g., to understand the model behaviors and to try out new models. Dealing with such artifacts and tasks is cumbersome and largely left to the users. This paper describes our vision and implementation of a data and lifecycle management system for deep learning. First, we generalize model exploration and model enumeration queries from commonly conducted tasks by deep learning modelers, and propose a high-level domain specific language (DSL), inspired by SQL, to raise the abstraction level and thereby accelerate the modeling process. To manage the variety of data artifacts, especially the large amount of checkpointed float parameters, we design a novel model versioning system (dlv), and a read-optimized parameter archival storage system (PAS) that minimizes storage footprint and accelerates query workloads with minimal loss of accuracy. PAS archives versioned models using deltas in a multi-resolution fashion by separately storing the less significant bits, and features a novel progressive query (inference) evaluation algorithm. Third, we develop efficient algorithms for archiving versioned models using deltas under co-retrieval constraints. We conduct extensive experiments over several real datasets from computer vision domain to show the efficiency of the proposed techniques.
more »
« less
ModelHub: Deep Learning Lifecycle Management
Deep learning has improved the state-of-the-art results in many domains, leading to the development of several systems for facilitating deep learning. Current systems, however, mainly focus on model building and training phases, while the issues of lifecycle management are largely ignored. Deep learning modeling lifecycle contains a rich set of artifacts and frequently conducted tasks, dealing with them is cumbersome and left to the users. To address these issues in a comprehensive manner, we demonstrate ModelHub, which includes a novel model versioning system (dlv), a domain-specific language for searching through model space (DQL), and a hosted service (ModelHub).
more »
« less
- Award ID(s):
- 1650755
- NSF-PAR ID:
- 10041785
- Date Published:
- Journal Name:
- Data Engineering (ICDE), 2017 IEEE 33rd International Conference on
- Page Range / eLocation ID:
- 1393 to 1394
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Deep learning has become the most popular direction in machine learning and artificial intelligence. However, the preparation of training data, as well as model training, are often time-consuming and become the bottleneck of the end-to-end machine learning lifecycle. Reusing models for inferring a dataset can avoid the costs of retraining. However, when there are multiple candidate models, it is challenging to discover the right model for reuse. Although there exist a number of model-sharing platforms such as ModelDB, TensorFlow Hub, PyTorch Hub, and DLHub, most of these systems require model uploaders to manually specify the details of each model and model downloaders to screen keyword search results for selecting a model. We are lacking a highly productive model search tool that selects models for deployment without the need for any manual inspection and/or labeled data from the target domain. This paper proposes multiple model search strategies including various similarity-based approaches and non-similarity-based approaches. We design, implement and evaluate these approaches on multiple model inference scenarios, including activity recognition, image recognition, text classification, natural language processing, and entity matching. The experimental evaluation showed that our proposed asymmetric similarity-based measurement, adaptivity, outperformed symmetric similarity-based measurements and non-similarity-based measurements in most of the workloads.more » « less
-
Obeid, I. ; Selesnik, I. ; Picone, J. (Ed.)The Neuronix high-performance computing cluster allows us to conduct extensive machine learning experiments on big data [1]. This heterogeneous cluster uses innovative scheduling technology, Slurm [2], that manages a network of CPUs and graphics processing units (GPUs). The GPU farm consists of a variety of processors ranging from low-end consumer grade devices such as the Nvidia GTX 970 to higher-end devices such as the GeForce RTX 2080. These GPUs are essential to our research since they allow extremely compute-intensive deep learning tasks to be executed on massive data resources such as the TUH EEG Corpus [2]. We use TensorFlow [3] as the core machine learning library for our deep learning systems, and routinely employ multiple GPUs to accelerate the training process. Reproducible results are essential to machine learning research. Reproducibility in this context means the ability to replicate an existing experiment – performance metrics such as error rates should be identical and floating-point calculations should match closely. Three examples of ways we typically expect an experiment to be replicable are: (1) The same job run on the same processor should produce the same results each time it is run. (2) A job run on a CPU and GPU should produce identical results. (3) A job should produce comparable results if the data is presented in a different order. System optimization requires an ability to directly compare error rates for algorithms evaluated under comparable operating conditions. However, it is a difficult task to exactly reproduce the results for large, complex deep learning systems that often require more than a trillion calculations per experiment [5]. This is a fairly well-known issue and one we will explore in this poster. Researchers must be able to replicate results on a specific data set to establish the integrity of an implementation. They can then use that implementation as a baseline for comparison purposes. A lack of reproducibility makes it very difficult to debug algorithms and validate changes to the system. Equally important, since many results in deep learning research are dependent on the order in which the system is exposed to the data, the specific processors used, and even the order in which those processors are accessed, it becomes a challenging problem to compare two algorithms since each system must be individually optimized for a specific data set or processor. This is extremely time-consuming for algorithm research in which a single run often taxes a computing environment to its limits. Well-known techniques such as cross-validation [5,6] can be used to mitigate these effects, but this is also computationally expensive. These issues are further compounded by the fact that most deep learning algorithms are susceptible to the way computational noise propagates through the system. GPUs are particularly notorious for this because, in a clustered environment, it becomes more difficult to control which processors are used at various points in time. Another equally frustrating issue is that upgrades to the deep learning package, such as the transition from TensorFlow v1.9 to v1.13, can also result in large fluctuations in error rates when re-running the same experiment. Since TensorFlow is constantly updating functions to support GPU use, maintaining an historical archive of experimental results that can be used to calibrate algorithm research is quite a challenge. This makes it very difficult to optimize the system or select the best configurations. The overall impact of all of these issues described above is significant as error rates can fluctuate by as much as 25% due to these types of computational issues. Cross-validation is one technique used to mitigate this, but that is expensive since you need to do multiple runs over the data, which further taxes a computing infrastructure already running at max capacity. GPUs are preferred when training a large network since these systems train at least two orders of magnitude faster than CPUs [7]. Large-scale experiments are simply not feasible without using GPUs. However, there is a tradeoff to gain this performance. Since all our GPUs use the NVIDIA CUDA® Deep Neural Network library (cuDNN) [8], a GPU-accelerated library of primitives for deep neural networks, it adds an element of randomness into the experiment. When a GPU is used to train a network in TensorFlow, it automatically searches for a cuDNN implementation. NVIDIA’s cuDNN implementation provides algorithms that increase the performance and help the model train quicker, but they are non-deterministic algorithms [9,10]. Since our networks have many complex layers, there is no easy way to avoid this randomness. Instead of comparing each epoch, we compare the average performance of the experiment because it gives us a hint of how our model is performing per experiment, and if the changes we make are efficient. In this poster, we will discuss a variety of issues related to reproducibility and introduce ways we mitigate these effects. For example, TensorFlow uses a random number generator (RNG) which is not seeded by default. TensorFlow determines the initialization point and how certain functions execute using the RNG. The solution for this is seeding all the necessary components before training the model. This forces TensorFlow to use the same initialization point and sets how certain layers work (e.g., dropout layers). However, seeding all the RNGs will not guarantee a controlled experiment. Other variables can affect the outcome of the experiment such as training using GPUs, allowing multi-threading on CPUs, using certain layers, etc. To mitigate our problems with reproducibility, we first make sure that the data is processed in the same order during training. Therefore, we save the data from the last experiment and to make sure the newer experiment follows the same order. If we allow the data to be shuffled, it can affect the performance due to how the model was exposed to the data. We also specify the float data type to be 32-bit since Python defaults to 64-bit. We try to avoid using 64-bit precision because the numbers produced by a GPU can vary significantly depending on the GPU architecture [11-13]. Controlling precision somewhat reduces differences due to computational noise even though technically it increases the amount of computational noise. We are currently developing more advanced techniques for preserving the efficiency of our training process while also maintaining the ability to reproduce models. In our poster presentation we will demonstrate these issues using some novel visualization tools, present several examples of the extent to which these issues influence research results on electroencephalography (EEG) and digital pathology experiments and introduce new ways to manage such computational issues.more » « less
-
Co-creation in higher education is the process where students collaborate with instructors in designing the curriculum and associated educational material. This can take place in different scenarios, such as integrating co-creation into an ongoing course, modifying a previously taken course, or while creating a new course. In this Work-In-Progress, we investigate training and formative assessment models for preparing graduate students in engineering to participate as co-creators of educational material on an interdisciplinary topic. The topic of cyber-physical systems engineering and product lifecycle management with application to structural health monitoring is considered in this co-creation project. This entails not only topics from different disciplines of civil, computer, electrical and environmental engineering, business, and information sciences, but also humanistic issues of sustainability, environment, ethical and legal concerns in data-driven decision-making that support the control of cyber-physical systems. Aside from the objective of creating modules accessible to students with different levels of disciplinary knowledge, the goal of this research is to investigate if the co-creation process and the resulting modules also promote interest and engagement in interdisciplinary research. A literature survey of effective training approaches for co-creation and associated educational theories is summarized. For students, essential training components include providing (i) opportunities to align their interests, knowledge, skills, and values with the topic presented; (ii) experiential learning on the topic to help develop and enhance critical thinking and question posing skills, and (iii) safe spaces to reflect, voice their opinions, concerns, and suggestions. In this research we investigate the adaption of project-based learning (PjBL) strategies and practices to support (i) and (ii) and focus groups for participatory action research (PAR) as safe spaces for reflection, feedback, and action in item (iii). The co-creation process is assessed through qualitative analysis of data collected through the PjBL activities and PAR focus groups and other qualitative data (i.e., focus group transcripts, interview transcripts, project materials, fieldnotes, etc.). The eventual outcome of the co-creation process will be an on-line course module that is designed to be integrated in existing engineering graduate and undergraduate courses at four different institutions, which includes two state universities and two that are historically black colleges and universities.more » « less
-
In a software system’s development lifecycle, engineers make numerous design decisions that subsequently cause architectural change in the system. Previous studies have shown that, more often than not, these architectural changes are unintentional by-products of continual software maintenance tasks. The result of inadvertent architectural changes is accumulation of technical debt and deterioration of software quality. Despite their important implications, there is a relative shortage of techniques, tools, and empirical studies pertaining to architectural design decisions. In this paper, we take a step toward addressing that scarcity by using the information in the issue and code repositories of open-source software systems to investigate the cause and frequency of such architectural design decisions. Furthermore, building on these results, we develop a predictive model that is able to identify the architectural significance of newly submitted issues, thereby helping engineers to prevent the adverse effects of architectural decay. The results of this study are based on the analysis of 21,062 issues affecting 301 versions of 5 large open-source systems for which the code changes and issues were publicly accessible.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1109/ICDE.2017.192
