Special Session at CBMI 2023
20-22 September 2023
Orleans, France
https://cbmi2023.org

Despite the advances in automated content description using deep learning, and the emergence of joint image-text embedding models, many video retrieval tasks still require a human user in the loop. This is in particular the case when the information need is fuzzy, or when the underlying dataset is homogeneous, i.e. contains data from one domain, differing in small details and with little or no editorial structure. Interactive video retrieval (IVR) systems address these challenges. In order to assess their performance, multimedia retrieval benchmarks such as Video Browser Showdown (VBS) or Lifelog Search Challenge (LSC) have been established. These benchmarks provide large-scale datasets as well as task settings and evaluation protocols, allowing to measure progress in research on IVR systems. However, in order to achieve the best possible performance of the participating systems, they are usually operated by members of the development team. This setting does not allow to pro!
 perly measure usability aspects of the system, which are important in order to deploy them successfully in a target application context, where they need to be operated by domain experts rather than video retrieval researchers.

This special session thus aims at providing better insights into how such systems are usable by users with solid IT background, but not familiar with the details behind the system. The special session thus calls for papers describing IVR systems, addressing topics such as

– search functionalities supporting non-expert users
– browsing and navigation capabilities
– approaches to result visualisation
– usability aspects of the system

The contributions to this session are short papers (4 pages + references), describing the participating IVR system. In contrast to papers such as submitted to VBS or LSC, these papers should focus on how the system supports users that are not retrieval experts, and the search and browsing features expected to be of particular interest to these users.

The review process is single-blind. A link to a three-minute video showcasing the usage of the system on the VBS collection must be included with the submission. Prior participation on VBS is not a prerequisite for submission.

The IVR systems will be presented in the demo session, followed by an interactive competition session, in which the systems are used by novice users to solve video retrieval tasks.

Important dates

Paper submission: April 12, 2023
Notification of acceptance: June 1, 2023
Camera ready paper: June 15, 2023
Conference dates: September 20-22, 2023

Organisers

Werner Bailer, JOANNEUM RESEARCH, Graz, Austria (werner.bailer@joanneum.at)
Cathal Gurrin, DCU, Ireland (cathal.gurrin@dcu.ie)
Björn Þór Jónsson, Reykjavík University, Iceland (bjorn@ru.is)
Klaus Schöffmann, Klagenfurt University, Austria (Klaus.Schoeffmann@aau.at)

https://2023.scadl.org

Scope of the Workshop:
Recently, Deep Learning (DL) has received tremendous attention in the research
community because of the impressive results obtained for a large number of
machine learning problems. The success of state-of-the-art deep learning
systems relies on training deep neural networks over a massive amount of
training data, which typically requires a large-scale distributed computing
infrastructure to run. In order to run these jobs in a scalable and efficient
manner, on cloud infrastructure or dedicated HPC systems, several interesting
research topics have emerged which are specific to DL. The sheer size and
complexity of deep learning models when trained over a large amount of data
makes them harder to converge in a reasonable amount of time. It demands
advancement along multiple research directions such as, model/data
parallelism, model/data compression, distributed optimization algorithms for
DL convergence, synchronization strategies, efficient communication and
specific hardware acceleration.

SCADL seeks to advance the following research directions:
– Asynchronous and Communication-Efficient SGD: Stochastic gradient descent is
at the core of large-scale machine learning. Parallelizing SGD gradient
computation across multiple nodes increases the data processed per iteration,
but exposes the SGD to communication and synchronization delays and
unpredictable node failures in the system. Thus, there is a critical need to
design robust and scalable distributed SGD methods to achieve fast error-
convergence in spite of such system variabilities.
High performance computing aspects: Deep learning is highly compute intensive.
Algorithms for kernel computations on commonly used accelerators (e.g. GPUs),
efficient techniques for communicating gradients and loading data from storage
are critical for training performance.

– Model and Gradient Compression Techniques: Techniques such as reducing
weights and the size of weight tensors help in reducing the compute
complexity. Using lower-bit representations such as quantization and
sparsification allow for more optimal use of memory and communication
bandwidth.

– Distributed Trustworthy AI: New techniques are needed to meet the goal of
global trustworthiness (e.g., fairness and adversarial robustness) efficiently
in a distributed DL setting.

– Emerging AI hardware Accelerators: with the proliferation of new hardware
accelerators for AI such in memory computing (Analog AI) and neuromorphic
computing, novel methods and algorithms need to be introduced to adapt to the
underlying properties of the new hardware (example: the non-idealities of the
phase-change memory (PCM) and the cycle-to-cycle statistical variations).

– The intersection of Distributed DL and Neural Architecture Search (NAS): NAS
is increasingly being used to automate the synthesis of neural networks.
However, given the huge computational demands of NAS, distributed DL is
critical to make NAS computationally tractable (e.g., differentiable
distributed NAS).

This intersection of distributed/parallel computing and deep learning is
becoming critical and demands specific attention to address the above topics
which some of the broader forums may not be able to provide. The aim of this
workshop is to foster collaboration among researchers from distributed/
parallel computing and deep learning communities to share the relevant topics
as well as results of the current approaches lying at the intersection of
these areas.

Areas of Interest
In this workshop, we solicit research papers focused on distributed deep
learning aiming to achieve efficiency and scalability for deep learning jobs
over distributed and parallel systems. Papers focusing both on algorithms as
well as systems are welcome. We invite authors to submit papers on topics
including but not limited to:

– Deep learning on cloud platforms, HPC systems, and edge devices
– Model-parallel and data-parallel techniques
– Asynchronous SGD for Training DNNs
– Communication-Efficient Training of DNNs
– Scalable and distributed graph neural networks, Sampling techniques for
graph neural networks
– Federated deep learning, both horizontal and vertical, and its challenges
– Model/data/gradient compression
– Learning in Resource constrained environments
– Coding Techniques for Straggler Mitigation
– Elasticity for deep learning jobs/spot market enablement
– Hyper-parameter tuning for deep learning jobs
– Hardware Acceleration for Deep Learning including digital and analog
accelerators
– Scalability of deep learning jobs on large clusters
– Deep learning on heterogeneous infrastructure
– Efficient and Scalable Inference
– Data storage/access in shared networks for deep learning
– Communication-efficient distributed fair and adversarially robust learning
– Distributed learning techniques applied to speed up neural architecture
search

Workshop Format:
Due to the continuing impact of COVID-19, ScaDL 2023 will also adopt relevant
IPDPS 2023 policies on virtual participation and presentation. Consequently,
the organizers are currently planning a hybrid (in-person and virtual) event.

Submission Link:
Submissions will be managed through linklings. Submission link available at:
https://2023.scadl.org/call-for-papers

Key Dates
Paper Submission:February 13th, 2023 (final deadline)
Acceptance Notification: February 17th, 2023
Camera ready papers due: February 28th, 2023
Workshop Date: TBA

Author Instructions
ScaDL 2023 accepts submissions in two categories:
– Regular papers: 8-10 pages
– Short papers/Work in progress: 4 pages
The aforementioned lengths include all technical content, references and
appendices.
We encourage submissions that are original research work, work in progress,
case studies, vision papers, and industrial experience papers.
Papers should be formatted using IEEE conference style, including figures,
tables, and references. The IEEE conference style templates for MS Word and
LaTeX provided by IEEE eXpress Conference Publishing are available for
download. See the latest versions at
https://www.ieee.org/conferences/publishing/templates.html

General Chairs
Kaoutar El Maghraoui, IBM Research AI, USA
Daniele Lezzi, Barcelona Supercomputing Center, Spain

Program Committee Chairs
Misbah Mubarak, NVIDIA, USA
Alex Gittens, Rensselaer Polytechnic Institute (RPI), USA

Publicity Chairs
Federica Filippini, Politecnico di Milano, Italy
Hadjer Benmeziane, Université Polytechnique des Hauts-de-France

Web Chair
Praveen Venkateswaran, IBM Research AI, USA

Steering Committee
Parijat Dube, IBM Research AI, USA
Vinod Muthusamy, IBM Research AI, USA
Ashish Verma, IBM Research AI, USA
Jayaram K. R., IBM Research AI, USA
Yogish Sabharwal, IBM Research AI, India
Danilo Ardagna, Politecnico di Milano, Italy