In Memory of Svatopluk Poljak
photo.gif,

Svata had a very broad interest and a wide mathematical knowledge. This is best illustrated by his many important contributions in diverse areas of discrete mathematics and combinatorial optimization such as matroid theory, matching theory, the max-cut and stable set problem, spectral approaches to graph problems, convex and polyhedral relaxations, semidefinite programming, and various other integer programming related problems. The fast approximation algorithms for finding a maximum cut by semidefinite programming, which are currently in the spotlight and form a major part of this volume, find their root in work of Svata on eigenvalue methods for graph problems. In addition, Svata's early work included pioneering contributions to the area of neural networks.

Svata was involved in a fatal car accident. This occurred 62 kms from Prague on Sunday evening April 2, 1995. Svata was on his way back to Prague from his summer house in Nove Hute. This is a village in the mountains, very conveniently located about halfway between Passau and Prague. Svata had just bought the cottage and he was really thrilled with it. In winter you can go skiing, starting right from the front door. It was one of Svata's longstanding dreams to have a place like that ... He was in the car with Standa, one of his best friends from Prague, two women including Standa's wife, and two children. They were hit by a car that came the other way, but on their side. There was no way to escape. Svata and his friend must have been dead instantly. The children have now largely recovered from their physical injuries, but the two women are still undergoing treatment.

Svata and Standa belonged to a group of four very good friends, who were engaged in fencing from the age of 10. The four of them and, later also their families, have remained inseparable throughout the years. They shared, among other things, a common love for hiking, canoeing, cross-country skiing, mountain trips, biking, etc...

This tragic accident happened at a moment which may have been among the happiest in Svata's life. Svata was indeed very happy with his new professional activity in Passau; he liked very much his work and the university environment, as well as living in Passau with his family. During the winter of 1995, he enjoyed fully the snowy weekends in Nove Hute, where he could happily mix work and leisure.

Svata will remain a point of reference in our community in many respects. We regret deeply the loss of a colleague and friend.

Many friends and colleagues sent letters of sympathy to Jana. (In particular, Jana thanks Monique Laurent and Franz Rendl for their moral support and help following the accident.) Jana has collaborated in choosing several anecdotes that follow.

• from Michele Conforti:
  I did not know Svata too well, but there is a story I would like to tell. A few years ago, Svata applied for a position and he asked me for a letter of recommendation. He sent me his vitae together with some publications. When I looked into his stuff, it became obvious that he was a much more advanced and established researcher than I was. I felt quite embarrassed to write a letter for such a strong (and versatile!) person. I think that he had a much larger mathematical culture than the average researcher in our area, but his low profile played (although won him some admirers, like me) a bit against him. When he proposed strikingly innovative research, as the use of semidefinite programming to obtain bounds for integer programs, he had trouble getting it accepted. It is very sad that only now we are here to recognize the value of this very honest person.
• from Francois Jaeger:
  I always enjoyed to be with Svata; to share with him the love of mathematics and the love of mountains. He was so enthusiastic, so full of life and so committed to everything he undertook. When we talked about mathematics, he was never superficial. He always wanted to go as deep as possible, either to understand fully some part of my knowledge,or to explain to me in the greatest detail some part of his. When we went hiking, he always wanted to see more, to see what was on the other side of the ridge. He was brave and would happily engage in adventurous steps without proper experience and equipment. I also liked his frankness and honesty, and his lucid and open minded vision of our job and our society........
• from Monique Laurent:
  I first met Svata in Paris in 1988 at the occasion of the Conference Coding and Counting. We worked on the metric polytope, we walked around Montmartre, and so started a strong friendship and fruitful collaboration. This friendship was soon extended to Svata's family. Jana and Svata always offered a warm hospitality to friends and colleagues, which was very dear to me, most preciously in Bonn, where we were simultaneously visiting in 1991/92. Our last encounter was in Amsterdam last November. This was a very fruitful and exciting month, during which we explored geometric questions related to positive semidefinite programming. Svata was a warm and kind person, very generous and helpful, also very shy, modest, and vulnerable. He was clever, always full of ideas and eager to share them. Svata's patience and enthusiasm were a constant source of inspiration for me. I have learned a lot of mathematics from him, and I miss very much the constant exchange of ideas we had together.
• from Martin Loebl:
  My memory of Svata is closely associated with the Prague Combinatorial Seminar which we both attended; Svata as one of its pillars and I as a novice. I immediately liked Svata because of his singular friendliness, and we became close friends and later collaborators. The main topics of our joint work were totally unimodular matrices and factors of graphs. Let me mention one joint result I am particularly proud of: we classified the algorithmic complexity of the {K_2,H} factor problem, i.e.
  • let H be a fixed graph; for a given graph H decide if its vertices may be covered by vertex-disjoint edges and copies of H.
  Svata made fundamental contributions in many areas of combinatorial optimization, local optimization and positive-semidefinite programming, his last subject which is covered most by this special issue. Svata enjoyed life fully and doing mathematics played a major role in his life. He will be sorely missed.
• from Jarik Nesetril:
  Svata Poljak was my first diploma student which I supervised at Charles University. We had contacts ever since and Poljak , Turzik and my families and all our then small kids took regular summer canoeing trips on various Bohemian rivers. Some papers started there. And some 10 years later I was happy to get Svata to Charles University back. I didn't expected this to last so briefly.
• from Franz Rendl:
  My first personal meeting with Svata was at a conference on partitioning in Grottaferrata (Italy) in 1991. We realized very quickly that we had many common research interests. This lead to a fruitful cooperation, but perhaps more important, to a deep friendship. Many of our joint results were obtained in a rather relaxed environment, where we would combine family holidays with mathematical discussions. It is sad for me to see several of our papers appearing in print only after his death.
• from Henry Wolkowicz:
  I met Svata while on sabbatical leave at DIMACS. Svata was visiting at the same time and we were introduced by our mutual friend Franz Rendl. Svata immediately showed me a conjecture he was working on:
  • four, seemingly unrelated, bounds for the max-cut problem are in fact all equal.
  Though I knew little about max-cut problems, the bounds involved continuous optimization techniques, which fell into my area of expertise. My first impression of Svata is how he never lost patience while explaining the details of the conjecture to me. His enthusiasm was infectious and I joined in to work on this problem. Our joint effort resulted in a proof of this conjecture. This work started our collaboration. Svata next met me in Graz, Austria and ``took'' me by car to his new home in Passau. Ironically, he disliked driving very much and I did all the driving from Graz to Passau. At this time he was essentially commuting from his family home in Prague to Passau. My lasting memory of Svata is spending several extremely enjoyable days at Passau discussing Mathematics and helping a very excited Svata explore his new home.